E7050 in Combination With Cetuximab Versus Cetuximab Alone in the Treatment of Platinum-Resistant Squamous Cell Carcinoma of the Head and Neck: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for E7050 in Combination With Cetuximab Versus Cetuximab Alone in the Treatment of Platinum-Resistant Squamous Cell Carcinoma of the Head and Neck (NCT01332266), a randomized phase 1, phase 2 study. The primary outcome is Phase 1b: Number of Participants With Dose-limiting Toxicities (DLTs) as Per National Cancer Institute Common Terminology Criteria for Adverse Events Version 4.0 (NCI CTCAE v4.0).

2. Clinical Question

The clinical question therefore has a safety and dose-finding component in Phase 1b and a comparative treatment component in Phase 2. The Phase 1b endpoint records whether participants experienced specified toxicities within the defined cycle. It is not itself a measure of tumor response or longer-term disease control.

For Phase 2, the registry lists progression-free survival, progression-free survival at Week 12, time to progression, overall survival, and overall response among the secondary outcomes. The posted numerical estimates include the first three, but numerical results for overall survival and overall response are not reported in the registry record. The analysis should therefore distinguish the Phase 1b toxicity count from the Phase 2 outcome estimates rather than treating them as one endpoint or one analysis population.

3. Trial Design

This was an open-label, multicenter, randomized, Phase 1b/2 study with a parallel-group design and no masking. The registry reports total enrollment of 95 and a completed status. The recorded study start is 2011-09-19, and the primary completion date is 2016-01-31.

Phase 1b included three cohorts receiving different E7050 doses with cetuximab. The posted primary endpoint counts are 4 participants in Cohort 1, 5 in Cohort 2, and 3 in Cohort 3. No participants in these cohorts were recorded as having a dose-limiting toxicity for the primary endpoint. These small cohort counts make the toxicity findings descriptive; they do not by themselves establish how often an uncommon toxicity might occur in a broader population.

Phase 2 compared a combination arm with a cetuximab-alone arm. The posted outcome counts are 42 for the combination arm and 41 for cetuximab alone. The registry arm descriptions refer to an MTD combination in Phase 2. The posted results labels identify the combination as golvatinib 400 mg with cetuximab 400 mg/m^2, while the registry’s arm descriptions specify cetuximab 250 mg/m^2. This difference in the posted descriptions is relevant when interpreting the exact regimen represented.

The randomized comparison supports assessing outcomes across the assigned Phase 2 groups, but the open-label design means participants and study personnel were not masked to treatment assignment. The registry record does not report details of the randomization procedure, allocation concealment, stratification, or the Phase 2 analysis population. It also does not report a protocol-based sample-size justification. Total enrollment describes the full study and should not be mistaken for the number in either Phase 2 group or the number contributing to every outcome.

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: This study consists of two phases Phase 1b and Phase 2. Phase 1b: 12 participants were enrolled and received the study treatment; Phase 2: 83 participants were enrolled and received the study treatment.

Phase 1b

Reasons for not completing this period:

Phase 2

Reasons for not completing this period:

Analysis population for the primary outcome, as defined in the registry: Safety population was defined as all participants enrolled and randomized to treatment in study, except for those who drop out of the study prior to receiving any study treatment, or were without any safety assessments following the first dose of study treatment.

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: Safety population was defined as all participants enrolled and randomized into Phase 1b and Phase 2, except for those who drop out of study prior to receiving any study treatment, or were without any safety assessments following the first dose of study treatment.

6. Primary Endpoint

The registry lists the primary outcome as Phase 1b: Number of Participants With Dose-limiting Toxicities (DLTs) as Per National Cancer Institute Common Terminology Criteria for Adverse Events Version 4.0 (NCI CTCAE v4.0), measured over Cycle 1 (Cycle length is equal to [=] 28 days).

The registered primary endpoint is the number of Phase 1b participants with dose-limiting toxicities during Cycle 1, whose length is 28 days. A dose-limiting toxicity is defined by specified adverse events meeting severity and timing criteria within the cycle. The definition includes selected non-hematologic events, serious blood-count-related toxicities, study-drug-related death, and other toxicities judged to be dose-limiting by the dose escalation committee.

This type of endpoint is used in early-phase research to characterize serious or treatment-limiting toxicity at studied doses. Counting participants who meet a defined toxicity rule gives a direct, clinically interpretable summary of the early treatment period. It is different from counting every adverse event: a participant may have an adverse event without meeting the specific dose-limiting definition.

The registry reports zero participants with a primary-endpoint DLT in each of the three Phase 1b cohorts. Those counts describe only the participants assessed in those cohorts and the specified Cycle 1 window. The small cohort sizes mean the observed zero counts do not demonstrate that the risk of such events is zero. Nor does this Phase 1b endpoint answer the Phase 2 comparative question about disease outcomes. The Phase 2 progression outcomes are reported separately as secondary endpoints, and their estimates should not be treated as extensions of the Phase 1b DLT count.

The registry defines a safety population for this endpoint as enrolled and randomized participants, except those who left before receiving any treatment or had no safety assessments after the first dose. The record does not report whether any participants were excluded under those criteria or provide a separate endpoint-specific denominator beyond the posted cohort counts.

7. Why No Re-analysis Is Shown

8. Statistical Interpretation

9. Statistical Methods Explained

Several different statistical summaries appear in the registry record. They address different questions, so their values should not be read as interchangeable evidence.

What does a confidence interval express?

A confidence interval describes the uncertainty around an estimated quantity under the statistical method used. A relatively broad interval means the data allow a broad range of plausible values. For the Phase 2 hazard ratios, the reported 95% confidence intervals include 1. In a ratio measure, 1 is the value corresponding to equal event rates between groups under the measure’s model. Thus, the intervals do not isolate a clear difference in one direction. The registry does not report the specific method used to calculate these hazard-ratio intervals.

For the Week 12 risk difference, the interval ranges from -0.311 to 0.101. A risk difference of zero represents equal proportions in the two groups; this interval includes zero. The point estimate is one possible summary of the observed group contrast, but the interval conveys the uncertainty around it. The unadjusted p-value of 0.319 is another summary of the data under a no-difference reference, not a measure of the probability that either treatment is effective.

How should medians and hazard ratios be read?

A median time-to-event estimate is the time by which half of the relevant event-time distribution has experienced the event, subject to the analysis method and available follow-up. The reported medians for progression-free survival are both 15.71 weeks. Equal medians do not establish that the full event-time distributions are identical: a median reduces the distribution to one point and does not describe every feature of its timing.

A hazard ratio compares event rates over follow-up through a time-to-event model. It is not a difference between the two medians and should not be interpreted as a fixed difference in weeks. A value of 0.89 for progression-free survival and 1.03 for time to progression are point estimates; their confidence intervals show that the estimates are uncertain. The registry record does not report the model specification, assumptions, handling of censoring, or statistical method underlying these hazard ratios.

What do the unadjusted calculations add?

The Week 12 recalculation is an unadjusted comparison derived from the posted group counts. It does not account for covariates, stratification, or any other adjustment. Its risk difference, interval, and p-value are therefore descriptive comparative summaries based on those counts, not substitutes for a sponsor-reported analysis. The record lists no reported comparative analysis for this endpoint. Because the recalculation is unadjusted, it should be interpreted with that limitation in mind.

Planning illustrations based on observed effects are also not observed trial outcomes. They show how the number of participants per arm for 80% power or 90% power would vary under the stated assumptions and effect fractions. For time-to-event outcomes, the planning figures refer to total events rather than participants per arm. Such illustrations use the observed effect as an input; they do not establish that the study had a particular power before it began, and they are not evidence that an effect was detected. The registry does not report its own sample-size justification, so the illustrations should not be attributed to the original trial design.

Why can over-dispersion matter?

For count outcomes, over-dispersion means the observed variability exceeds what a simple count model would expect. If a method assumes less variability than the data contain, its uncertainty estimates can be too small and its intervals too narrow. This is a general statistical consideration, not a claim that over-dispersion was assessed or found in this study. The primary endpoint is reported as participant counts, and the registry does not report a model-based analysis of those counts.

With small cohorts, event counts may change substantially if only a few outcomes differ. A zero count is a useful description of what was observed, but it does not mean the underlying probability is known to be zero. The registry gives no confidence intervals for the Phase 1b DLT counts.

What does the event-time follow-up description mean?

Progression-free survival is defined from randomization to the earlier of progression or death; time to progression is defined from randomization to progression. These endpoints differ in which events count as outcomes. The registry lists a possible observation period of up to approximately 4 years 4 months for each. That description is not a statement that every participant was followed for the full period. The record does not report the number censored, the distribution of follow-up, or whether the reported estimates use a particular survival-curve estimator.

These distinctions matter because the medians, hazard ratios, confidence intervals, and Week 12 percentages summarize different aspects of the data. No single one of them fully describes all participant outcomes, and the registry does not provide enough methodological detail to reconstruct every reported analysis.

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("Phase 1b: Cohort 1: Golvatinib 200 mg + Cetuximab ", "Phase 1b: Cohort 2: Golvatinib 300 mg + Cetuximab ", "Phase 1b: Cohort 3: Golvatinib 400 mg + Cetuximab ")
n      <- c(4, 5, 3)
events <- c(0, 0, 0)
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  Phase 1b: Cohort 1: Golvatinib 200 mg + Cetuximab   4  0  0.0  0.0  60.2  0.0  49.0  0.5  52.2
ARM  Phase 1b: Cohort 2: Golvatinib 300 mg + Cetuximab   5  0  0.0  0.0  52.2  0.0  43.4  0.4  45.9
ARM  Phase 1b: Cohort 3: Golvatinib 400 mg + Cetuximab   3  0  0.0  0.0  70.8  0.0  56.1  0.6  60.2

11. Multiplicity, Missing Data and Analysis Populations

The registry record does not report how multiplicity across the secondary outcomes was handled. It lists progression-free survival, progression-free survival at Week 12, time to progression, overall survival, and overall response, but does not describe an adjustment procedure, testing hierarchy, or family of hypotheses. When several endpoints are examined, the chance of at least one apparently unusual result can increase. Without a reported strategy, nominal or unadjusted results should not be interpreted as if they were protected against that issue.

Missing-data handling is also not reported in the registry record for the Phase 2 outcomes. The record does not describe how participants with missing assessments, incomplete follow-up, or unavailable endpoint information were handled. For time-to-event analyses, it does not report censoring rules or the number censored. These details can affect both the estimates and their precision. The listed numbers of participants in the Phase 2 arms should not automatically be assumed to be the denominator for every endpoint at every time point.

The record defines a safety population for the Phase 1b DLT endpoint: enrolled and randomized participants are included except those who left before receiving any study treatment or had no safety assessments after the first dose. It does not report how many participants, if any, were excluded by these rules. The record does not report a corresponding Phase 2 analysis-population definition, such as whether the posted comparisons follow randomized assignment or require treatment receipt.

Details of interim analyses and any adjustments related to them are not reported in the registry record. Nor are the statistical analysis plan, the handling of protocol deviations, or the extent to which analyses followed the original plan. These absences limit assessment of how completely the posted analyses reflect the intended inferential framework. They do not establish that a particular procedure was or was not used; they mean the registry record does not document it.

12. Safety Results

The registry reports serious adverse events as affected participants divided by participants at risk.

Deaths

Other (non-serious) adverse events

Participants with at least one non-serious event at or above the registry's 5% reporting threshold.

Most frequent serious adverse event terms

Most frequent non-serious adverse event terms

Serious events with the highest arm-level rate: pneumonia (33.3%), pneumonia aspiration (33.3%), upper airway obstruction (25.0%), abdominal pain (20.0%), nausea (20.0%).

The serious adverse event figures are counts of affected participants among those listed at risk in each group. In Phase 1b, serious adverse events were reported for 1 participant in Cohort 1, 4 in Cohort 2, and 2 in Cohort 3. In Phase 2, the corresponding counts were 15 in the combination arm and 15 in the cetuximab-alone arm. These counts should be read alongside the different group sizes and the study’s safety definitions; they are not, by themselves, adjusted comparisons of risk.

Other adverse events were recorded for all 4 participants in Cohort 1, all 5 in Cohort 2, all 3 in Cohort 3, all 42 participants in the Phase 2 combination arm, and 38 in the cetuximab-alone arm. The death counts were 3, 3, and 2 across the Phase 1b cohorts, and 33 and 36 in the Phase 2 arms. The registry figures do not establish that deaths were caused by study treatment.

The listed serious-event terms and percentages are descriptive. The registry does not provide enough information in these figures to infer causality, duration, severity distribution, or comparative safety after adjustment. A participant may have more than one event, so event terms should not be treated as mutually exclusive participant categories.

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

The posted results establish what was recorded in the registry: no Phase 1b participants in the listed cohorts met the primary DLT endpoint during Cycle 1, and the Phase 2 estimates have substantial uncertainty. The Phase 2 progression-free survival medians were equal as reported, while the hazard-ratio confidence interval includes values on either side of 1. The Week 12 comparison is an unadjusted recalculation whose interval includes no difference. The time-to-progression hazard-ratio interval likewise includes 1.

These estimates do not establish that the two Phase 2 groups had identical outcomes. An interval that includes the no-difference value can reflect uncertainty and limited precision; it does not prove equivalence. Conversely, a point estimate alone does not establish a reliable advantage. The registry record does not report a prespecified equivalence or noninferiority framework, so the results should not be interpreted as demonstrating either.

The Phase 1b DLT counts do not establish the safety profile for all participants or for longer periods. They concern a narrowly defined endpoint, a short stated assessment window, and small cohorts. The broader adverse-event counts provide additional descriptive context but do not establish that an event was caused by treatment.

The record also leaves important questions unanswered. It does not report a numerical Phase 2 result for overall survival or overall response, the statistical analysis plan, missing-data procedures, or the sample-size justification. Differences between the registry arm descriptions and the posted results labels for cetuximab dose add uncertainty about how the regimen is represented. Taken together, the registry supports a careful account of observed counts and estimates, not a definitive claim about comparative clinical benefit or harm.

14. Why This Trial Matters Statistically

This study illustrates why a trial’s endpoint, phase, and analysis population need to be kept distinct. Its registered primary endpoint is a Phase 1b DLT count, whereas the numerical comparative outcome estimates come from Phase 2 secondary endpoints. A single description of “the result” would obscure that the toxicity endpoint and the disease-outcome comparisons concern different questions and groups.

It also illustrates the importance of precision. The Phase 2 point estimates are accompanied by confidence intervals that permit a wide range of comparative effects. Reading only the medians or point estimates would understate that uncertainty. The Week 12 calculation adds a direct unadjusted group contrast, but it is not a sponsor-reported analysis and its interval also leaves room for effects in either direction.

The registry’s limited reporting of analysis methods is statistically important. Without documented details on multiplicity, missing data, censoring, or the Phase 2 analysis population, readers cannot fully assess how the estimates were produced or how robust they are to analytic choices. The planning illustrations based on observed effects are useful as hypothetical design calculations, but they cannot stand in for the trial’s own design rationale or be read as observed power.

Finally, the small Phase 1b cohorts demonstrate why zero observed DLTs should be described as a count rather than a guarantee of no risk. The study’s record is most informative when its estimates, uncertainty, and reporting gaps are considered together.

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 trial affect interpretation. First, the primary endpoint is reported for small Phase 1b cohorts, with no confidence intervals or model-based comparison. Zero DLTs in these cohorts provide limited information about less common events and do not characterize toxicity beyond the defined Cycle 1 window.

Second, the Phase 2 estimates are imprecise. The confidence intervals for both reported hazard ratios include 1, and the Week 12 risk-difference interval includes zero. The registry record does not provide p-values for the reported hazard-ratio analyses. The Week 12 p-value is an unadjusted recalculation, not a sponsor-reported result. Numerical Phase 2 estimates for overall survival and overall response are not reported.

Third, key analysis details are absent. The record does not report the Phase 2 analysis population, missing-data rules, censoring details, model specifications, multiplicity handling, or interim analyses. It also does not report the trial’s own sample-size justification. Consequently, the planning illustrations based on observed effects should not be used to infer the design’s original assumptions or adequacy.

Fourth, the regimen descriptions differ: the registry arm descriptions specify cetuximab 250 mg/m^2, while the posted results labels specify cetuximab 400 mg/m^2. The combination is also labeled E7050 in the study title and golvatinib in the posted results. The registry record does not explain these differences, so the precise relationship between the descriptions is not reported.

Finally, the open-label design means treatment assignment was not masked. The record does not report whether outcome assessment or other procedures addressed possible influence of that design. These limitations do not erase the posted observations, but they constrain how far those observations can support broader comparative conclusions.

16. Related Tutorials

17. Related Statistical Calculators

18. Sources

• ClinicalTrials.gov: E7050 in Combination With Cetuximab Versus Cetuximab Alone in the Treatment of Platinum-Resistant Squamous Cell Carcinoma of the Head and Neck — NCT01332266.
• Registry record last updated 2022-01-03; status last verified 2017-12.
• Results first posted 2022-01-03.
• 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

The registry describes a completed, open-label, multicenter, randomized, parallel Phase 1b/2 study of E7050 with cetuximab versus cetuximab alone for platinum-resistant squamous cell carcinoma of the head and neck. Total enrollment was 95. Phase 1b’s primary endpoint was the number of participants with DLTs during Cycle 1; the posted counts were zero in each of the three cohorts.

Phase 2 posted progression-free survival, Week 12 progression-free survival, and time-to-progression estimates for 42 participants in the combination arm and 41 in the cetuximab-alone arm. The reported confidence intervals are broad and include the respective no-difference values. Numerical results for overall survival and overall response are not reported in the registry record. Key analysis details, including multiplicity and missing-data methods, are also not reported.