Phase II Randomized Trial of the Combination of Cetuximab and Sorafenib or Single Agent Cetuximab: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Phase II Randomized Trial of the Combination of Cetuximab and Sorafenib or Single Agent Cetuximab (NCT00939627), a randomized phase 2 study. The primary outcome is Progression Free Survival (PFS).

2. Clinical Question

3. Trial Design

This was a randomized, open-label, parallel-group phase II trial sponsored by the National Cancer Institute (NCI). The registry lists enrollment as 55 and the status as completed. The start date is 2009-07, and the primary completion date is 2013-01. The registry does not report a masking procedure; its design description specifies no masking.

One arm received cetuximab with placebo, and the other received cetuximab with sorafenib tosylate. Cetuximab was administered intravenously on days 1, 8, and 15. The oral study treatment, whether placebo or sorafenib tosylate, was scheduled twice daily on days 1-21. Laboratory biomarker analysis and quality-of-life assessment are also listed as interventions or procedures.

The primary outcome was progression-free survival, defined from the on-study date to the earlier of disease progression or death from any cause. The registry describes progression using RECIST v1.1: an increase of at least 20% in the sum of longest diameters of target lesions, unequivocal progression of non-target lesions, or appearance of new lesions. The outcome was assessed up to 3 years, using the Kaplan-Meier method.

The posted arm records list n=26 for each arm's primary outcome. The safety records list 26 at risk in the cetuximab arm and 27 at risk in the combination arm. The reason these arm-level denominators differ from one another and from the registry's enrollment figure is not reported in the registry record. Randomization details, such as the method used to generate the allocation sequence, 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.

Before assignment: Fifty-five patients were consented and assigned to a treatment arm; 3 withdrew before treatment. Fifty-two patients were treated on study. After 19 patients were enrolled, the trial was amended to remove the placebo (and blinding) due to issues with placebo tablet solubility.

Overall Study

Reasons for not completing this period:

Analysis population for the primary outcome, as defined in the registry: All patients are included in the analysis on intention-to-treat basis. Analysis is by Kaplan-Meier method, where either death or progression is an event, with censoring for non-progressed, non-expired patients at greater of off-study date or last known alive date.

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: Baseline participant measures are based on total eligible, consented, patients. Three of 55 consenting patients did not receive treatment. One patient had disease progression before treatment and 2 patients withdrew before treatment, resulting in 52 patients on treatment.

6. Primary Endpoint

The registry lists the primary outcome as Progression Free Survival (PFS), measured over On-study date to lesser of date of progression or date of death from any cause (assessed up to 3 years).

Progression-free survival measures the time from the on-study date until disease progression or death from any cause, whichever occurs first. It is a time-to-event outcome: each participant contributes information about how long they remain alive without recorded progression, subject to the trial's event and censoring rules. The registry defines progression according to RECIST v1.1 and specifies that the endpoint was assessed up to 3 years.

The reported statistic is the median progression-free survival estimated by the Kaplan-Meier method, with a 95% confidence interval. A median describes the time by which the estimated progression-free survival curve reaches its midpoint. The confidence interval communicates uncertainty around that arm-specific estimate; it is not itself a comparison between arms.

This endpoint is used to describe the timing of progression or death rather than simply counting how many participants experienced an event by a particular visit. The registry counts either progression or death as an event and censors participants who have neither. Censoring allows participants with incomplete event follow-up to contribute the time they were observed, but interpretation depends on the recorded follow-up and censoring approach. The registry specifies censoring for non-progressed, non-expired patients at the greater of the off-study date or last known alive date.

7. Why No Re-analysis Is Shown

8. Statistical Interpretation

9. Statistical Methods Explained

What does a Kaplan-Meier estimate represent?

Kaplan-Meier analysis estimates the proportion of participants who remain free of the event over time while accounting for participants whose event time is not observed during follow-up. In this trial, progression or death from any cause counts as an event. A participant who has not had either event by the relevant censoring time contributes observed follow-up up to that point, rather than being treated as if an event occurred then.

This approach is useful for time-to-event outcomes because follow-up may end at different times for different participants. It uses event timing and available follow-up rather than reducing the outcome to a simple event count. Its interpretation depends on the event definitions, dates, and censoring rules being applied as specified. The registry describes the censoring rule for non-progressed, non-expired patients, but does not report further details about follow-up completeness or censoring patterns.

What does the median mean?

The median progression-free survival is the time at which the estimated Kaplan-Meier curve reaches the midpoint of its survival distribution. It is a summary of the curve, not the average time without progression or death and not a statement about every participant. Participants may have event times both earlier and later than the median. A median can be useful for a plain-language description, but it does not show the full shape of the curve or whether the relative difference between arms changes over time.

The posted medians are 3.0 months for cetuximab and 3.2 months for cetuximab with sorafenib tosylate. Each is accompanied by its own confidence interval. Those intervals describe uncertainty in the individual estimates; they are not a confidence interval for the difference between arms. A direct contrast requires an appropriate comparative estimate and its uncertainty, which are not reported in the registry record.

How should the confidence intervals be read?

A confidence interval conveys the statistical uncertainty associated with an estimate under the analysis method. A narrower interval generally indicates greater precision, while a wider interval indicates less precision. Here, the intervals are 1.9 to 5.0 months for cetuximab and 1.8 to 4.2 months for the combination. They should be read as uncertainty around the respective medians, not as a range containing most individual participants' event times and not as a treatment-effect interval.

Confidence intervals do not remove the need to assess the study design, analysis population, and data completeness. They also do not, by themselves, establish whether a difference is statistically significant. A comparison would typically require a reported treatment-effect estimate and uncertainty measure, or a stated test of the contrast. Such a result is not reported in the registry record.

Why does censoring matter?

Censoring occurs when a participant's event time is not observed before follow-up ends. That participant's available event-free time still informs the Kaplan-Meier estimate, but the exact later event time is unknown. The method relies on assumptions about how censored observations relate to the event process. The registry describes when censoring is applied, but does not report diagnostic assessments of those assumptions or detailed censoring counts.

For this endpoint, the registry states that non-progressed, non-expired patients are censored at the greater of the off-study date or last known alive date. The record does not provide a participant-level account of censoring or reasons for ending follow-up. Consequently, the posted summary cannot show whether censoring differed between arms or how much that might affect the curve estimates.

What can be concluded from the posted analysis?

The primary endpoint is identified as a median time-to-event measure, and the registry gives arm-level estimates and confidence intervals. The record does not give a hazard ratio or another comparative analysis for this endpoint. The trial record also indicates that no analyses are posted for the comparison, and notes that values are missing or not estimable for a two-arm comparison of a median time endpoint. A median alone does not supply the full information needed to estimate a relative treatment effect.

Sample-size calculations, when used as planning illustrations, are based on assumptions and are not observed power. The registry does not report the trial's sample-size justification or the assumptions behind its original design calculations. Any planning illustration should therefore be kept distinct from the trial's observed results and should not be interpreted as proof that the study had adequate or inadequate power.

10. Multiplicity, Missing Data and Analysis Populations

The registry lists secondary outcomes of best response, overall survival, number of participants with each worst-grade toxicity, gene expression levels, and overall survival associated with immunomodulatory cytokines. It lists seven outcome measures, but numerical results for these secondary outcomes are not reported in the record. The statistical plan for testing multiple outcomes, including any adjustment for multiplicity or hierarchy among endpoints, is not reported in the registry record. Without that information, the chance of false-positive findings across a set of tests cannot be assessed.

The primary outcome description specifies an intention-to-treat analysis and Kaplan-Meier estimation. It does not provide a detailed flow of participants through analysis, explain exclusions or missing endpoint assessments, or report the amount and reasons for missing follow-up. The endpoint description states that non-progressed, non-expired participants are censored at the greater of off-study date or last known alive date. This is an analysis rule, but the record does not provide enough detail to assess how censoring was distributed between arms.

The registry enrollment is 55, while the arm-level primary outcome records list n=26 in each arm and the safety denominators are 26 and 27. An explanation of these denominator differences is not reported in the registry record. They should not be interpreted as evidence of particular exclusions, missing outcomes, or a specific analysis set without participant disposition information.

Information on interim analyses, missing-data sensitivity analyses, and prespecified handling of secondary outcomes is not reported in the registry record. Accordingly, the posted primary summaries can be described, but the full analysis strategy and robustness of results to missing data or multiplicity cannot be evaluated.

11. 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: death nos (11.1%), dysphagia (7.7%), dehydration (7.4%), anaphylaxis (3.8%), neck pain (3.8%).

The safety record reports serious adverse events in 5 of 26 participants at risk in the cetuximab arm and 10 of 27 participants at risk in the cetuximab-and-sorafenib arm. These are counts of participants affected in the registry's serious-event summary. They are not rates per unit of follow-up, and the record does not provide a comparative statistical test or uncertainty interval for the difference.

For other adverse events, the record lists 25 of 26 participants affected in the cetuximab arm and 27 of 27 in the combination arm. The registry does not provide further detail in this summary about event duration, severity distribution, or exposure time. Safety counts can be influenced by follow-up, treatment exposure, and event ascertainment, so the figures alone do not establish a causal difference between arms.

The registry lists serious-event terms and percentages: Death NOS, 11.1%; Dysphagia, 7.7%; Dehydration, 7.4%; Anaphylaxis, 3.8%; Neck pain, 3.8%; and Neoplasms benign, malignant and unspecified (incl cysts and polyps), 3.8%. The record does not specify arm attribution for these listed percentages in the summary. Deaths by arm are not reported in the registry record. These terms should therefore be read as reported safety descriptions, not as an arm-specific comparison or an estimate of treatment causation.

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

The posted results describe progression-free survival in the two randomized arms using Kaplan-Meier medians and confidence intervals. The median was 3.0 months (95% confidence interval 1.9 to 5.0) with cetuximab and 3.2 months (95% confidence interval 1.8 to 4.2) with cetuximab plus sorafenib tosylate. Those estimates characterize the trial record's endpoint summaries; they do not give a direct estimate of the treatment contrast.

The record does not report a hazard ratio, a confidence interval for a comparative effect, or a p-value for the primary endpoint. Therefore, the posted summaries do not establish the direction or precision of a formal between-arm effect. Nor do they establish that the treatments are equivalent. A difference between medians, viewed without a comparative analysis, is not enough to make those conclusions.

The randomized design is relevant to the intended comparison, but interpretation also depends on participant follow-up, outcome assessment, and analysis details. The record states that the primary analysis was intention-to-treat, yet does not explain the discrepancy between the registry enrollment and arm-level denominators. It also does not report detailed missing-data or censoring information beyond the stated censoring rule.

Safety summaries report serious and other adverse-event counts, but no comparative statistical analysis is reported. The listed serious-event terms are not attributed to arms in the record's summary, and deaths by arm are not reported. These data should not be treated as proof that one regimen caused more or fewer events.

Overall, the registry supports a descriptive account of the posted arm-level endpoint estimates and safety counts. It does not, by itself, provide enough comparative statistical information to determine the size or precision of a treatment effect. It also does not report numerical results for the listed secondary outcomes in this record.

13. Why This Trial Matters Statistically

This trial illustrates the distinction between a randomized comparison and a complete report of a comparative treatment effect. Random allocation is part of the design, but the posted primary results are medians with separate confidence intervals for each arm. Those summaries describe the outcome within each arm; they are not a direct estimate of the contrast created by randomization.

Time-to-event outcomes require attention to both event timing and censoring. Progression-free survival combines progression and death into one endpoint and uses follow-up from participants who have not yet experienced either event. The Kaplan-Meier method can incorporate incomplete follow-up, but interpretation still depends on how events and censoring are defined and recorded. The registry gives the main rules, while leaving censoring patterns and participant disposition unreported.

The record also demonstrates why denominators matter. Enrollment is listed as 55, while the primary outcome and safety summaries use different arm-level denominators. Without an explanation, readers cannot determine how those records relate to participant flow or whether the analysis populations differ. Reporting that reconciliation would support more transparent interpretation.

Finally, a list of secondary outcomes does not substitute for their results, and the absence of a reported comparative estimate limits what can be concluded from the primary endpoint. Statistical interpretation needs the estimate, its uncertainty, and enough information about the analysis to understand how it was obtained. These distinctions help keep descriptive results separate from claims about comparative efficacy or safety.

14. 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 most immediate limitation of the posted primary results is that they give arm-specific medians and confidence intervals but not a comparative treatment-effect estimate. A hazard ratio, confidence interval for the contrast, and p-value are not reported in the registry record. As a result, the magnitude and statistical precision of the between-arm comparison cannot be assessed from the posted primary summaries.

The registry enrollment figure is 55, whereas the primary endpoint arm records list n=26 in each arm, and the safety risk denominators are 26 and 27. The reason for these differences is not reported in the registry record. Participant flow, exclusions, and the relation between enrollment and each analysis denominator are therefore unclear.

The registry describes an intention-to-treat analysis and the censoring rule, but does not report detailed follow-up completeness, censoring counts, or reasons for missing outcome information. This limits assessment of how much incomplete follow-up may influence the Kaplan-Meier estimates. The record also does not report the randomization sequence method, allocation concealment, or additional analysis details.

Numerical results for the listed secondary outcomes are not included in the record summarized. Multiplicity handling and interim analyses are not reported in the registry record. Safety results are limited to participant counts and a list of serious-event terms; the listed terms are not assigned to arms in the summary, and deaths by arm are not reported. These gaps restrict the scope of conclusions that can be drawn from the posted record.

15. Related Tutorials

16. Related Statistical Calculators

17. Sources

• ClinicalTrials.gov: Phase II Randomized Trial of the Combination of Cetuximab and Sorafenib or Single Agent Cetuximab — NCT00939627.
• Registry record last updated 2026-03-10; status last verified 2017-12.
• Results first posted 2015-09-01.
• 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.

18. Record Summary

NCT00939627 is a completed, randomized, unmasked, parallel-group phase II trial sponsored by the National Cancer Institute (NCI). It compared cetuximab with placebo against cetuximab with sorafenib tosylate in patients with refractory, recurrent, and/or metastatic squamous cell carcinoma of the head and neck. The registry lists enrollment as 55, with a start date of 2009-07 and primary completion date of 2013-01.

The primary outcome was progression-free survival, measured from the on-study date to progression or death from any cause and estimated using Kaplan-Meier methods. The reported medians were 3.0 months (95% confidence interval 1.9 to 5.0) for cetuximab and 3.2 months (95% confidence interval 1.8 to 4.2) for the combination. A comparative effect estimate is not reported in the registry record.

Serious adverse events affected 5 of 26 participants at risk in the cetuximab arm and 10 of 27 in the combination arm. The record lists secondary outcomes, but numerical results for them are not reported in the registry record.