Expansion Trial for Axitinib In Head And Neck Cancer: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Expansion Trial for Axitinib In Head And Neck Cancer (NCT02762513), a na phase 2 study. The primary outcome is Number of Evaluable Patients Alive at 6 Months.

2. Clinical Question

In statistical terms, this is a descriptive question about survival status in the trial’s defined evaluable population. A result for this endpoint could describe what was observed among participants who received at least one cycle, but it would not by itself estimate a treatment effect against a control. The registry record also lists median overall survival, median progression-free survival, best overall response, and grade 3 or worse toxicities as secondary outcomes. Their numerical results are not reported in the registry record, so the available information does not establish their observed values or precision.

3. Trial Design

The registry describes a phase II, single-group trial of axitinib in patients with unresectable recurrent or metastatic head and neck squamous cell carcinoma. The study is listed as completed, with enrollment of 29. The sponsor is the University of Michigan Rogel Cancer Center. The record gives a start date of 2016-08-30 and a primary completion date of 2020-04-09. The brief summary describes the study as prospective and single-institution.

Participants were to start axitinib at 5 mg twice a day continuously, with escalation to 7 mg and then 10 mg twice a day if grade 2 or worse toxicities were absent. Clinical and/or radiologic response assessment was described after 8 weeks and subsequently every 2 months until disease progression or intolerable toxicity. These details describe the registered treatment and assessment approach; the registry does not report how treatment or assessment unfolded for each participant.

The study had no randomized allocation and no masking. With no concurrent comparator, the design can describe outcomes observed in the treated group but cannot isolate an effect of axitinib from other explanations for those outcomes. The registry does not report a protocol-based sample-size justification, interim analyses, or a detailed statistical analysis plan. Those features should not be inferred from the phase label, the enrollment figure, or the presence of a primary endpoint.

The primary endpoint is defined among evaluable participants, meaning those who received at least one cycle. This population definition is narrower than simply referring to all participants enrolled, and the number of evaluable participants is not reported in the registry record. The distinction matters when reading any percentage: its denominator should match the endpoint’s stated population, not be assumed from the enrollment figure.

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: 1 patient enrolled but progressed before starting study treatment.

Overall Study

5. Baseline Characteristics

The table reproduces the baseline characteristics posted in the registry, which describe the participants analysed at baseline and help show whether the randomized groups were comparable.

6. Primary Endpoint

The registry lists the primary outcome as Number of Evaluable Patients Alive at 6 Months, measured over 6 Months after treatment initiation.

The primary endpoint is the number and percentage of evaluable participants alive 6 months after treatment initiation. Evaluable participants are defined in the registry as those who received at least one cycle of axitinib. The endpoint records whether each person is alive at a specified follow-up time; it is not a measure of the duration of survival across the full follow-up period.

A fixed-time survival proportion can be useful because it expresses survival status at a clinically specified landmark in a way that is straightforward to describe. It is distinct from median overall survival, which summarizes the timing of deaths across follow-up, and from progression-free survival, which concerns time until progression or death under the endpoint definition. The registry lists both median overall survival and median progression-free survival as secondary outcomes, but their numerical results are not reported in the registry record.

The registry specifies a 95% confidence interval as the endpoint’s dispersion measure. In general, a confidence interval conveys the statistical uncertainty around an estimated proportion: a wider interval indicates less precision, while a narrower interval indicates greater precision. The observed endpoint count, percentage, and interval are not reported in the registry record, so neither the estimate nor its precision can be assessed from the available outcome entry.

Interpretation also depends on complete and consistent survival ascertainment at the landmark and on a clearly defined evaluable population. The registry record does not report the number assessed at 6 months, the handling of missing survival status, or the approach used to calculate the confidence interval.

7. Why No Re-analysis Is Shown

8. Statistical Interpretation

9. Statistical Methods Explained

Statistical interpretation begins by distinguishing the type of outcome from the design used to observe it. The primary endpoint is a binary status at a fixed time: each evaluable participant is classified as alive or not alive at 6 months after treatment initiation. A percentage expresses the number with the specified status relative to the endpoint’s defined population. The registry’s definition of evaluable is receipt of at least one cycle of axitinib. The count and percentage themselves are not reported in the registry record.

What does a confidence interval add?

In general, an estimate from a sample is subject to sampling variation. A confidence interval gives a range of values compatible with the data under the method used and communicates uncertainty around the estimate. It is not a guarantee that the true value lies within the particular interval, and it does not remove limitations caused by the study design or incomplete outcome information. The registry specifies a 95% confidence interval for the primary endpoint, but the interval and calculation method are not reported in the registry record. Its width, and therefore the precision of the estimate, cannot be assessed.

Why does the single-group design matter?

A single-group study estimates outcomes among participants treated in that group. It does not create a randomized counterfactual group with which to compare those outcomes. Randomization, when used, helps balance measured and unmeasured prognostic factors on average between groups; it is not present in this trial. A single-group estimate can still be descriptive, but it cannot establish a comparative treatment effect on its own. The record lists allocation as not applicable and the model as single group.

Masking is listed as none. In general, masking can reduce the influence of knowledge about treatment assignment on assessments or decisions, depending on the outcome and study procedures. In a study with no masking, outcome assessment may be more vulnerable to observer or participant expectations. The registry does not describe the assessment procedures in enough detail to determine whether, or how, this could have affected the endpoints.

How should fixed-time survival be distinguished from time-to-event outcomes?

A fixed-time status asks whether a participant is alive at the specified landmark. A time-to-event outcome uses both the timing of an event and the time during which a participant was observed without that event. Median overall survival and median progression-free survival are listed as secondary outcomes, but their results and analysis methods are not reported in the registry record. In general, time-to-event analyses need to address censoring, which occurs when an event time is not observed during a participant’s available follow-up. The registry does not report its censoring approach.

General statistical methods for proportions may also depend on the number observed and on assumptions about how observations vary. The registry does not state the method used to calculate the primary confidence interval, so a particular interval procedure should not be attributed to the trial. If a proportion is based on a small or uncertain denominator, its interval can be important for showing that the point estimate alone may give an incomplete impression of precision. Here, the denominator and interval are not reported in the registry record.

What do the planning illustrations mean?

Sample-size calculations are planning exercises, not outcomes from the completed trial. The planning approach described for the associated illustration uses a two-sided significance level of 0.05, equal allocation between two arms, and a normal approximation; the effect to detect is expressed as a fraction of the observed effect. Time-to-event calculations are expressed as a total number of events, while other calculations are expressed as participants per arm. Those assumptions describe the illustration, not the registered single-group design. The registry does not report that this was the trial’s original sample-size plan. Such calculations are not observed power, and they do not show that a study was capable of detecting a particular effect after the fact.

For a proportion outcome, over-dispersion is a general concern when the variability in observations exceeds what a simple model assumes. It can make a conventional interval too narrow if the model understates variability. Whether over-dispersion was assessed or addressed in this trial is not reported in the registry record. More broadly, statistical precision and causal interpretation are separate: a narrow interval around a single-group proportion would not create a controlled comparison, while a broad interval would signal uncertainty about the group’s estimate.

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("Axitinib")
n      <- c(28)
events <- c(20)
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  Axitinib  28  20  71.4  51.3  86.8  52.9  84.7  52.8  84.7

11. Multiplicity, Missing Data and Analysis Populations

The registry lists a primary endpoint and several secondary outcomes: median overall survival, median progression-free survival, best overall response, and grade 3 or worse toxicities. It also records 5 outcome measures and 2 analyses. The record does not report whether the analyses were confirmatory or exploratory, how outcome testing was ordered, or whether any adjustment for multiple comparisons was planned.

In general, testing many outcomes can increase the chance of finding at least one apparently notable result by chance. Statistical approaches to multiplicity can control that risk, but the suitable approach depends on the study’s objectives and analysis plan. No such approach is reported in the registry record, so no multiplicity adjustment should be assumed. The numerical results for the listed efficacy outcomes are also not reported in the registry record.

Missing outcome information is another distinct issue. The primary endpoint is defined in an evaluable population, but the number evaluable and the number with survival status determined at 6 months are not reported in the registry record. The record does not describe whether missing status was excluded, imputed, or handled by another method. These details affect how the denominator and resulting percentage should be interpreted.

The registry does not report the analysis populations used for the secondary outcomes, the handling of treatment discontinuation, or the approach to incomplete follow-up. The safety entry identifies participants at risk for adverse events, but that does not establish the population used for each efficacy analysis. Without these details, the scope of each analysis and the potential influence of missing observations remain unclear.

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 0% reporting threshold.

Most frequent serious adverse event terms

Most frequent non-serious adverse event terms

Serious events with the highest arm-level rate: diarrhea (7.1%), sepsis (7.1%), cardiac arrest (3.6%), abdominal pain (3.6%), cholecystitis (3.6%).

The safety entry lists 28 participants at risk for serious adverse events and 8 affected. For other adverse events, it lists 28 at risk and 27 affected. It also records 20 deaths. These are registry safety counts; they do not establish that an event or death was caused by axitinib. The record does not provide a causal assessment in the information.

The serious-event terms listed most often include diarrhea and sepsis, each at 7.1%. Cardiac arrest, abdominal pain, cholecystitis, and infections and infestations - Other are each listed at 3.6%. These percentages should be read as the registry’s reported figures, not recalculated or treated as estimates of comparative risk. The safety entry does not supply a control group for comparison.

In general, an adverse-event count describes events recorded under the study’s monitoring and reporting procedures. It does not alone distinguish events caused by treatment from events related to the underlying illness, other care, or unrelated circumstances. The registry record does not report exposure time by dose, event timing, event-level overlap, or the number of participants reaching each dose. Thus, the counts and listed terms describe the safety entry but do not support a dose-specific or causal interpretation.

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

The registry record establishes that this was a completed phase II, single-group study of axitinib in people with unresectable recurrent or metastatic head and neck squamous cell carcinoma. It specifies a primary endpoint based on survival status 6 months after treatment initiation among participants who received at least one cycle. It also describes the planned starting dose and conditional dose escalation, the assessment schedule, and several secondary outcomes.

However, the record does not report the primary endpoint’s observed count, percentage, or confidence interval. It also does not report the numerical results for the listed secondary efficacy outcomes. Consequently, the available information cannot quantify the six-month survival proportion, its precision, overall or progression-free survival, or best overall response. No conclusion about whether axitinib did or did not improve an outcome follows from the missing numerical results.

Even if the study’s outcome estimates were available, the single-group design would limit causal interpretation. A result from participants receiving axitinib would describe their observed outcomes, not the difference between axitinib and a concurrent comparison treatment. There is no randomized or like-for-like control group in this design. Likewise, safety counts alone do not establish that axitinib caused the recorded events or deaths.

The safety entry provides counts for participants at risk and affected, a death count, and selected serious-event percentages. These details can be read as recorded safety information, but the entry does not provide a comparative risk estimate or enough information to attribute events. Missing denominator details for the primary endpoint and absent analysis methods also limit assessment of precision and possible bias.

In short, the registry supports a description of the trial’s design and some safety reporting, but it does not provide the numerical primary result needed to answer the stated six-month endpoint question. The limits of a single-group study remain relevant regardless of the size or precision of any observed proportion.

14. Why This Trial Matters Statistically

This trial illustrates a basic distinction in clinical-trial interpretation: an outcome definition is not the same as an outcome result. The registry specifies a fixed-time survival endpoint and identifies the population used to define it, but the observed count, percentage, and confidence interval are not reported in the registry record. Without those values, readers cannot judge either the result’s direction or its precision.

The design also highlights why a treatment group’s outcome is not automatically a treatment effect. A single-group estimate can be informative as a description of what occurred in the study population, but it does not supply a concurrent counterfactual. The lack of a like-for-like control group limits causal comparisons, even when an endpoint is clearly defined and follow-up is described.

The record further shows why denominators must be interpreted outcome by outcome. Enrollment is listed as 29, the primary endpoint is restricted to evaluable participants, and the safety entry lists 28 at risk. These figures belong to different contexts. Treating them as interchangeable would obscure what population each measure represents; the registry does not report the primary evaluable denominator.

Finally, the study’s dosing escalation condition and safety reporting demonstrate that treatment exposure and adverse-event summaries can involve more detail than a single group-level count. The registry does not report dose-level exposure or event timing, so those aspects cannot be examined. The statistical lesson is not that the available figures resolve efficacy or safety, but that careful interpretation depends on the design, endpoint population, uncertainty, and completeness of the reported results.

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.

The primary endpoint result is not reported in the registry record. The observed number alive at 6 months, the percentage, the 95% confidence interval, and the endpoint denominator are unavailable. The evaluable definition is stated, but the number of participants meeting it is not reported. The record also does not describe the handling of missing survival status, so the completeness and analysis population for the endpoint cannot be assessed.

The single-group design has no concurrent comparator and no randomized allocation. This limits the ability to distinguish treatment-associated outcomes from the effects of participant characteristics, disease course, other care, or other influences. Masking is listed as none; details about outcome adjudication or safeguards against assessment bias are not reported.

The registry does not report a sample-size justification, interim analyses, or a detailed statistical analysis plan. It does not specify the confidence-interval method, the analysis of time-to-event outcomes, the handling of censoring, or any multiplicity strategy. The numerical results of the secondary efficacy outcomes are not reported in the registry record, which prevents assessment of their estimates or precision.

The safety entry lists counts and selected serious-event terms, but does not report event timing, exposure duration, dose-specific safety, or causal attribution. It also does not establish a comparison group for safety. The enrollment figure and the safety at-risk count should not be taken as the primary endpoint denominator. These limitations are specific to the information recorded and do not establish that additional analyses or procedures were absent; they are not reported in the registry record.

16. Related Tutorials

17. Related Statistical Calculators

18. Sources

• ClinicalTrials.gov: Expansion Trial for Axitinib In Head And Neck Cancer — NCT02762513.
• Registry record last updated 2021-05-05; status last verified 2021-04.
• Results first posted 2021-05-05.
• 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 phase II, single-group study of axitinib for unresectable recurrent or metastatic head and neck squamous cell carcinoma. Enrollment is listed as 29. The study is sponsored by the University of Michigan Rogel Cancer Center, with a start date of 2016-08-30 and a primary completion date of 2020-04-09.

The primary endpoint is the number and percentage of evaluable participants alive 6 months after treatment initiation; evaluable participants are defined as those receiving at least one cycle. The observed endpoint result and its 95% confidence interval are not reported in the registry record. Numerical results for the listed secondary efficacy outcomes are also not reported.

The safety entry lists 28 at risk for serious adverse events, 8 affected, 28 at risk for other adverse events, 27 affected, and 20 deaths. Selected serious-event terms and their reported percentages are described in the safety discussion. These figures do not establish causation or provide a controlled comparison.