A Phase II Trial of Cetuximab and Bevacizumab in Patients With Recurrent or Metastatic Head and Neck Cancer: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for A Phase II Trial of Cetuximab and Bevacizumab in Patients With Recurrent or Metastatic Head and Neck Cancer (NCT00409565), a na phase 2 study. The primary outcome is Objective Response Rate (ORR).

2. Clinical Question

The study asks whether the combination was associated with tumor shrinkage or disappearance in people with recurrent or metastatic head and neck cancer. Objective response rate (ORR) is the primary endpoint. It classifies target-lesion changes as complete response, partial response, stable disease, or progressive disease, and counts complete and partial responses in the overall response category.

This is a single-group Phase II study, not a randomized comparison of the combination against another treatment. The registry record therefore does not support a direct estimate of how outcomes would differ under an alternative treatment or control condition. It reports that 45 of the 46 eligible participants were evaluable for response, but the actual ORR result is not reported in the registry record. Without that result, the central response question cannot be answered from the posted results. Other outcomes listed include progression-free survival, overall survival, change in serum cytokine concentrations, and disease control rate.

3. Trial Design

This completed study is identified as Phase II and single-group, with an enrollment of 48. The treatment group received cetuximab plus bevacizumab. Allocation is listed as not applicable, and masking as none. These features describe an open-label, nonrandomized design in which the registry record does not identify a concurrent control arm.

The study concerned recurrent or metastatic head and neck cancer, including squamous cell carcinoma. The listed start was 2006-09, and primary completion was 2012-02. The sponsor was the National Cancer Institute (NCI). The record states that 46 participants were eligible and 45 were evaluable for response. It does not explain why the enrolled, eligible, and response-evaluable counts differ.

A single-group design can describe outcomes observed among participants receiving the combination. It cannot, by itself, separate treatment-associated changes from the course of the disease, other care, or differences in who could be assessed. Without a concurrent comparator, between-group treatment effects cannot be estimated from this trial’s design. The registry record does not report further details about treatment schedules, eligibility criteria, follow-up procedures, or the statistical design rationale. Those details should not be inferred from the trial title or its Phase II designation.

4. Participant Flow and Accounting

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

Before assignment: We report results in 46 eligible patients. Two patients were deemed ineligible, one because subsequent biopsy showed that measurable disease was actually steoradionecrosis and not recurrent laryngeal cancer. The other patient required surgery for cholecystitis before starting treatment; this patient never initiated protocol treatment.

Overall Study

Analysis population for the primary outcome, as defined in the registry: Assess all patients included in this study for response to treatment, per protocol. Of 46 eligible patients, 45 were evaluable for response.

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: STARTED 46; Overall Response Rate 7; COMPLETED 45; Not Completed 1

6. Primary Endpoint

The registry lists the primary outcome as Objective Response Rate (ORR), measured over Up to 5 years.

The primary endpoint is objective response rate (ORR), defined as the percentage of participants whose cancer shrank or disappeared after study treatment. Response was assessed using RECIST criteria for target lesions, with computed tomography (CT) and/or magnetic resonance imaging (MRI). A complete response means all target lesions disappeared. A partial response requires at least a 30% decrease in the sum of the longest diameters of target lesions. Complete and partial responses together make up overall response.

The endpoint also distinguishes progressive disease from stable disease. Progressive disease requires at least a 20% increase in the sum of the longest diameters, using the smallest recorded sum since treatment began as the reference, or the appearance of one or more new lesions. Stable disease means the change meets neither the partial-response nor progressive-disease criteria.

As a general statistical concept, ORR turns a set of lesion assessments into a participant-level binary outcome: a participant either meets the overall-response definition or does not. It is used to describe how often tumor shrinkage meeting predefined criteria occurred. The registry record gives a timeframe of up to 5 years and states that 45 eligible participants were evaluable for response. The ORR estimate and its 95% confidence interval are not reported in the registry record.

7. Why No Re-analysis Is Shown

8. Statistical Interpretation

9. Statistical Methods Explained

The registry record describes a percentage-based primary endpoint, but does not report the actual ORR estimate, its confidence interval, or the method used to calculate that interval. The following statistical explanations are general; they do not identify an unreported method as the one used in this trial.

What does a response percentage represent?

For a binary endpoint such as objective response, each participant is classified according to whether the response definition is met. The response percentage describes the share of the analysis population classified as having an overall response. Its interpretation depends on the response criteria and on which participants are included in that population. In this record, overall response combines complete and partial response under the stated RECIST definitions. The record identifies 45 participants as evaluable for response, but it does not report the number who achieved either response category or the resulting percentage.

A percentage is a descriptive estimate, not a guarantee that the same proportion would be observed in another group of participants. As a general statistical concept, sampling variability means an estimate can vary across samples even when the underlying response probability is unchanged. A confidence interval is commonly used to convey uncertainty around a sample estimate. Its numerical limits and the method used to obtain them are not reported in the registry record, so precision cannot be assessed from the endpoint description alone.

How do the RECIST categories define the endpoint?

The endpoint relies on defined changes in target lesions. Complete response refers to disappearance of all target lesions. Partial response requires at least a 30% decrease in the sum of the longest diameters. Progressive disease is defined by at least a 20% increase from the smallest recorded sum since treatment began, or by one or more new lesions. Stable disease is the category for neither sufficient shrinkage for partial response nor sufficient increase for progressive disease. Overall response is the combination of complete and partial response. These rules provide a standardized classification framework, but the record does not provide participant-level measurements or response-category counts.

Imaging assessment by CT and/or MRI is specified. The registry record does not report further operational details, such as how assessments were timed, whether readings were centrally reviewed, or how ambiguous assessments were handled. Those practices can matter for reproducibility, but their presence or absence cannot be inferred.

What does the single-group design allow?

In general, a single-group trial can estimate outcomes among participants receiving the studied intervention, subject to the quality and completeness of outcome assessment. It does not create a randomized contrast with a control group. Thus, even if a response percentage were available, it would describe response in the observed group rather than a causal difference between treatment strategies. Differences in participant characteristics, disease course, additional care, or assessment can complicate comparisons with outcomes from other studies. The registry record does not report a prespecified external benchmark for interpreting ORR.

What do the planning assumptions mean?

The registry record describes planning illustrations based on a fraction of the observed effect, with a two-sided significance level of 0.05, equal allocation between two arms, and a normal approximation. Such calculations are planning exercises, not observed trial results or observed power. The stated two-arm allocation assumption does not describe the actual single-group design. A planning illustration also depends on an effect size; the record does not report the observed ORR estimate needed to understand the effect used in any such illustration. No sample-size justification for the trial itself is reported in the registry record.

As a general concept, power calculations estimate the sample size needed under specified assumptions to detect a chosen effect with a chosen probability, subject to the model. They do not show that an effect occurred, and they do not replace an estimate and confidence interval from observed data. Because this study has one treatment group, a two-arm planning framework should not be read as evidence that the trial actually randomized participants between arms.

What analysis details are unavailable?

The registry record does not report the statistical method for the primary endpoint, the confidence-interval procedure, a hypothesis test, an adjustment for covariates, or an interim-analysis plan. It also does not report whether response analyses followed an intention-to-treat approach or another prespecified analysis population. General methodological possibilities should not be mistaken for trial-specific details. The available information supports explaining the endpoint and design, but not reconstructing the analysis or recalculating an unreported result.

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("Cetuximab Plus Bevacizumab")
n      <- c(45)
events <- c(7)
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  Cetuximab Plus Bevacizumab  45  7  15.6  6.5  29.5  7.7  28.8  7.8  28.9

11. Multiplicity, Missing Data and Analysis Populations

The registry record names ORR as the primary outcome and lists progression-free survival (PFS), overall survival (OS), change in serum cytokine concentrations, and disease control rate (DCR), with clinical benefit rate also appearing in the DCR description, as secondary outcomes. It does not report how many hypotheses were tested, whether any multiplicity adjustment was used, or how primary and secondary findings were prioritized in the analysis. A multiplicity strategy is not reported in the registry record.

As a general statistical concept, when several outcomes or tests are examined, the chance of at least one apparently notable result can increase if each is interpreted separately without accounting for the collection of tests. Whether and how that issue was handled in this trial cannot be determined from the outcome names. The record does not report p-values or outcome estimates for the listed secondary outcomes.

The record says that 45 eligible participants were evaluable for response, compared with 46 eligible and 48 enrolled. It does not explain the differences, report reasons for unavailable response assessments, or describe imputation or other missing-data procedures. It also does not report the analysis populations used for the primary or secondary outcomes. The population statement says all patients included in the study were to be assessed for response per protocol, but the precise analysis rules and handling of participants without an evaluable response are not reported in the registry record.

Interim analyses, stopping rules, and any prespecified hierarchy for testing outcomes are not reported in the registry record. Their absence from the record should not be taken as evidence that they did or did not occur. These gaps limit interpretation of the completeness and inferential context of the posted results.

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: dysphagia (difficulty swallowing) (15.2%), anorexia (4.3%), dehydration (4.3%), pain, abdomen nos (4.3%), infection with grade 3 or 4 neutrophils, lung (pneumonia) (4.3%).

The registry record lists 46 participants at risk for serious adverse events and 22 affected participants in the cetuximab-plus-bevacizumab group. It lists 46 participants at risk for other adverse events and 46 affected. These are participant counts, not measures of how many separate events occurred, how long they lasted, or how severe every event was. The death count is not reported in the registry record.

The listed serious adverse-event terms and percentages are dysphagia (difficulty swallowing), 15.2%; anorexia, 4.3%; dehydration, 4.3%; pain, abdomen NOS, 4.3%; infection with Grade 3 or 4 neutrophils, lung (pneumonia), 4.3%; and trismus (difficulty, restriction or pain when opening mouth), 4.3%. The registry record does not give additional event-level detail for these terms. The percentages should be read as recorded, not recalculated or combined; participants may have experienced more than one event, and the categories are not stated to be mutually exclusive.

These figures describe events recorded in the study group. They do not, on their own, establish that the treatment caused an event or allow comparison with another group. The registry record does not provide a comparator group for safety interpretation.

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

The record documents a Phase II, single-group study of cetuximab plus bevacizumab in recurrent or metastatic head and neck cancer, with ORR designated as the primary outcome. It specifies response criteria and reports that 45 eligible participants were evaluable for response. It also lists serious and other adverse-event counts for the treatment group. These details describe the intended clinical question, the response framework, and some safety reporting.

However, the primary response estimate is not reported in the registry record. The response count, ORR percentage, and 95% confidence interval are not available in the posted results. Without them, the observed response rate and its precision cannot be described. Listing an endpoint or defining its criteria is not the same as reporting an outcome value.

The single-group design means the study does not provide a randomized comparison against another treatment or a control. Any reported outcome would describe the treated group, not a measured difference attributable to the combination relative to an alternative. The record does not report a comparative effect for efficacy or safety.

As a general statistical principle, an estimate’s interpretation depends both on how it was obtained and on uncertainty around it. Even a precise estimate in a single group would not resolve the absence of a concurrent comparator. Conversely, the absence of a reported estimate does not show that no response occurred; it means the result is unavailable in the registry record.

The safety counts likewise do not establish causation or comparative risk. They report affected participants in the study group, while the registry record does not provide a control group, death count, or further details needed to characterize event timing and duration. The information supports a description of the registered design and reported safety counts, but it does not support a conclusion about treatment benefit, lack of benefit, or relative safety.

14. Why This Trial Matters Statistically

This trial is statistically instructive because its primary endpoint is clearly defined while its primary result is not reported in the registry record. The distinction between defining an outcome and reporting its estimate is fundamental: response criteria tell readers how participants are classified, but they do not reveal how many met those criteria or how uncertain the resulting percentage is.

The design also illustrates why a single-group result has a different interpretation from a randomized comparison. A response proportion can describe what was observed in a treated cohort, but a causal comparison requires an appropriate counterfactual or control. The record identifies no randomized allocation or comparator group. Therefore, treatment-effect language should not be inferred solely from the existence of a response endpoint.

Participant counts matter for interpretation as well. The registry record gives 48 enrolled, 46 eligible, and 45 evaluable for response. These distinct counts indicate that the analysis population is not simply interchangeable with the enrollment total. Because the reasons for the differences and the missing-data handling are not reported, readers cannot assess whether exclusions or unavailable assessments might affect the estimate.

The safety information offers another useful distinction: affected-participant counts and listed event percentages are not interchangeable with event totals, causal attribution, or a comparison of risk. The absence of a comparator limits comparative safety interpretation just as it limits efficacy interpretation.

Finally, planning calculations should be separated from observed findings. Planning assumptions describe hypothetical requirements under specified conditions; they do not establish a result or observed power. The record’s two-arm planning assumptions do not alter the trial’s stated single-group design. Keeping endpoint definitions, observed results, uncertainty, and planning illustrations conceptually separate prevents stronger conclusions than the record supports.

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.

Several limitations specific to this record constrain statistical interpretation. First, the primary ORR result and its 95% confidence interval are not reported in the registry record. No response count, estimate, or precision measure is available to assess the primary outcome. The definition of response is detailed, but the outcome itself cannot be quantified from the posted results.

Second, the trial is single-group, with allocation listed as not applicable and masking as none. There is no randomized comparator in the registry record. Consequently, the design cannot provide a direct randomized estimate of the difference between cetuximab plus bevacizumab and another treatment.

Third, the record reports 48 enrolled, 46 eligible, and 45 evaluable for response without explaining the differences. The reasons participants were not eligible or evaluable, and the handling of missing response information, are not reported in the registry record. This prevents assessment of how the response analysis population was formed.

Fourth, the record does not report the statistical method for ORR, the confidence-interval procedure, hypothesis testing, covariate adjustment, interim analyses, multiplicity handling, or a sample-size justification for the trial. It also does not report estimates for the listed secondary outcomes. These omissions prevent reconstruction of the analysis plan and broader assessment of the evidence.

The safety record lists affected participants and selected serious-event terms, but the death count is not reported in the registry record. Additional details about event timing, duration, and attribution are also not reported. These limitations mean the available information supports a careful description of design and safety counts, not a complete assessment of efficacy or comparative safety.

16. Related Tutorials

17. Related Statistical Calculators

18. Sources

• ClinicalTrials.gov: A Phase II Trial of Cetuximab and Bevacizumab in Patients With Recurrent or Metastatic Head and Neck Cancer — NCT00409565.
• Registry record last updated 2017-10-20; status last verified 2017-09.
• Results first posted 2016-07-12.
• 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

This completed Phase II study evaluated cetuximab plus bevacizumab in a single group of people with recurrent or metastatic head and neck cancer. The registry record lists 48 enrolled, 46 eligible, and 45 evaluable for response. The primary outcome was ORR, assessed using RECIST criteria and CT and/or MRI, with a timeframe of up to 5 years.

The registry record does not report the ORR result, its 95% confidence interval, or results for the listed secondary outcomes. It lists 22 of 46 participants as affected by serious adverse events and 46 of 46 as affected by other adverse events; the death count is not reported. Because there was no comparator group, the record does not provide a randomized estimate of comparative efficacy or safety.