Trial of Ibrutinib Combined With Nivolumab or Cetuximab to Treat Recurrent/Metastatic HNSCC: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Trial of Ibrutinib Combined With Nivolumab or Cetuximab to Treat Recurrent/Metastatic HNSCC (NCT03646461), a randomized phase 2 study. The primary outcome is Clinical Efficacy of Combined Therapies Using RECIST v1.1.

2. Clinical Question

3. Trial Design

This was a randomized, open-label, parallel-group phase II trial sponsored by the University of California, San Diego. Participants were assigned to one of two active combinations: ibrutinib with cetuximab, or ibrutinib with nivolumab. The trial record describes no masking, so treatment assignment was not concealed from participants or investigators through blinding as recorded. The registry record does not describe the randomization sequence, allocation concealment, or the procedures used to implement assignment.

The stated enrollment was 5, matching the posted arm counts: 3 participants in the ibrutinib-plus-cetuximab arm and 2 in the ibrutinib-plus-nivolumab arm. The listed primary outcome was clinical efficacy based on best overall response under RECIST v1.1, assessed over up to 22 months. Progression-free survival and overall survival were listed as secondary outcomes. Numerical results for those secondary outcomes are not reported in the registry record.

The record lists a start date of 2018-10-17 and a primary completion date of 2020-08-15. The trial status is terminated. The reason for termination, the intended enrollment before termination, and whether recruitment or follow-up ended earlier than planned are not reported in the registry record. The record also does not report the planned duration of follow-up for each participant or the timing and frequency of tumor assessments.

Because the trial was randomized, its design was intended to compare outcomes across treatment assignments. However, very small arm counts limit what can be learned from the posted comparison. The registry record has no posted analysis records, so it does not show a formal estimate of the difference between combinations or a statistical test for that comparison. The arm-level counts can be described, but they do not establish that any apparent difference between treatments is real or attributable to treatment.

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

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 Clinical Efficacy of Combined Therapies Using RECIST v1.1, measured over Up to 22 months.

The primary endpoint was best overall response rate using RECIST v1.1. It counts participants whose best recorded tumor response was either a complete response or a partial response during the assessment period, which was listed as up to 22 months. A complete response means disappearance of all target lesions. A partial response is defined as a >=30% decrease in the sum of the longest diameters of target lesions. The overall response combines complete and partial responses.

This endpoint is used to describe tumor response in a defined and reproducible way. A response count is a clinically interpretable summary of how many participants met the stated imaging-based criteria, while RECIST provides rules for classifying changes in target lesions. Best overall response considers the most favorable qualifying response recorded over the assessment period, rather than only one isolated assessment. These are general properties of response endpoints; the registry record does not provide additional details about assessment schedules, confirmation requirements, or how missing assessments were treated.

The posted results report zero participants meeting the response endpoint in the ibrutinib-plus-cetuximab arm and zero in the ibrutinib-plus-nivolumab arm. The arm sizes were 3 and 2, respectively. These are counts from a very small enrolled group, not precise estimates of response in a broader patient population. The registry record does not report confidence intervals for the arm-level response results.

7. Why No Re-analysis Is Shown

8. Statistical Interpretation

9. Statistical Methods Explained

Several statistical ideas help explain what can and cannot be inferred from these results. The trial compared two active combinations in a randomized, parallel-group design. Its primary outcome was binary at the participant level: a participant either met the definition of complete or partial response, or did not. The posted arm results are counts of participants meeting that endpoint. This simple outcome format is easy to describe, but precision depends strongly on how many participants contribute information.

Why small counts limit precision

In a binary-outcome trial, each participant contributes an outcome classification. When only a few participants are enrolled, a change in the result for one person can materially change the observed pattern. This is a general statistical property, not evidence that any particular participant was misclassified. Here, the arm sizes are 3 and 2, and the posted response count is zero in both groups. The registry record does not report confidence intervals around these results. A confidence interval is a range of values compatible with the data under a specified statistical method; it communicates uncertainty rather than guaranteeing that the true population value lies in the interval. With no such interval reported, the uncertainty is not quantified in the registry results.

What randomization contributes

Random assignment is intended to make the treatment groups comparable, on average, with respect to factors that could influence the outcome. That purpose does not ensure that small groups will be balanced on every characteristic. Nor does randomization remove uncertainty caused by having few participants. The record identifies the trial as randomized but does not report the sequence-generation method, allocation concealment, or other implementation details. Those details therefore cannot be assessed from the registry record.

Why the posted counts do not constitute a formal comparison

Describing the two response counts is different from estimating and testing a treatment contrast. A formal comparison would typically specify an effect measure and an uncertainty estimate, with an analysis method appropriate to the design and outcome. The registry record lists 0 analyses. It does not report an effect estimate, confidence interval, or p-value for the primary comparison. It would not be appropriate to invent such a comparison from the counts or to treat the absence of a reported test as proof that the arms are the same.

The record identifies one arm as the control for registry purposes, but both study arms received active combination regimens. This does not make one combination a no-treatment control. The clinical comparison is between ibrutinib plus cetuximab and ibrutinib plus nivolumab. No single-agent or placebo arm is listed.

Response criteria and measurement

RECIST v1.1 supplies operational rules for classifying tumor response. The primary outcome uses the best overall response, combining complete and partial responses. The registry description defines complete response as disappearance of all target lesions and partial response as a >=30% decrease in the sum of the longest diameters of target lesions. Clear outcome definitions can support consistent classification across participants and centers. However, the registry record does not report the assessment schedule, whether responses required confirmation, who reviewed the assessments, or how assessments that could not be made were handled. Those omissions leave important measurement details unclear.

Over-dispersion and binary outcomes

Over-dispersion is a general concern when observed variability exceeds what a simple statistical model assumes. It is often discussed for count outcomes or clustered data, where outcomes within a group may be more alike than a basic model allows. The primary endpoint is reported as participant counts with a binary response, and the registry record does not describe a model or analysis that accounts for over-dispersion. No claim about over-dispersion in these results can be made from the posted information. More generally, with very small groups, complex model assumptions cannot be meaningfully evaluated from the reported counts alone.

Planning illustrations are not observed power

Any sample-size planning figures associated with these results are illustrations based on the observed effect and stated assumptions, rather than the trial's own design justification. The listed assumptions are a two-sided significance level of 0.05, equal allocation between arms, and a normal approximation; the effect to detect is described as a stated fraction of the observed effect. Such calculations explore how a hypothetical future design might behave under those assumptions. They are not evidence that the completed trial had a particular power, and they do not explain the enrollment target or the decision to terminate. The trial's own sample-size justification is not reported in the registry record.

For general interpretation, power is a design property related to the chance of detecting a specified effect under stated assumptions. It is not a measure of whether the observed result is true, and a planning illustration is not observed power. The registry record does not report a prespecified effect size, a sample-size calculation for the original trial, or an analysis plan that would permit checking how the completed study was intended to address its primary question.

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("Arm A: Ibrutinib + Cetuximab", "Arm B: Ibrutinib + Nivolumab")
n      <- c(3, 2)
events <- c(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  Arm A: Ibrutinib + Cetuximab  3  0  0.0  0.0  70.8  0.0  56.1  0.6  60.2
ARM  Arm B: Ibrutinib + Nivolumab  2  0  0.0  0.0  84.2  0.0  65.8  0.8  70.8

11. Multiplicity, Missing Data and Analysis Populations

The registry record names a primary response endpoint and lists progression-free survival and overall survival as secondary outcomes. Considering several outcomes can create a multiplicity issue: the more statistical comparisons made, the greater the chance that at least one result appears notable by chance if each is interpreted separately. Statistical plans may address this through prespecified testing order or adjustment, but the registry record does not report whether any such approach was planned or used. It also lists 0 analyses, and numerical results for the two secondary outcomes are not reported in the registry record.

The record does not describe whether response was tested separately within each arm, whether a between-arm comparison was prespecified, or which result would have priority if multiple outcomes were considered. It does not report interim analyses or stopping boundaries. These details are not available for interpretation; their absence from the registry record should not be read as evidence that a particular procedure was or was not used outside the posted record.

Missing-data handling and analysis populations are also not reported in the registry record. It is not stated how participants without an evaluable tumor assessment would be classified, whether any participants were excluded from the primary analysis, or whether analyses followed an intention-to-treat or other population definition. The posted enrollment and arm counts do not resolve those questions. The registry record also does not describe the completeness of follow-up for the primary outcome, progression-free survival, or overall survival.

These gaps matter because the interpretation of a binary response count can depend on who is included and how an unavailable response assessment is treated. They also prevent a reader from evaluating whether the registered endpoint, analysis population, and reported results correspond to a prespecified statistical plan.

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 non-serious adverse event terms

The registry record reports serious adverse events for 0 participants in the ibrutinib-plus-cetuximab arm, among 3 at risk, and 0 participants in the ibrutinib-plus-nivolumab arm, among 2 at risk. It reports other adverse events for 3 participants in the cetuximab arm and 2 in the nivolumab arm. Deaths are reported for 3 participants in the cetuximab arm and 2 in the nivolumab arm. These are counts, not estimates of comparative risk.

The record does not list serious adverse-event terms, and it does not report whether the deaths were considered related to treatment, disease, or another cause. The timing of adverse events and deaths, the criteria used to classify events, and their duration are not reported in the registry record. The counts alone therefore do not establish cause, severity beyond the recorded categories, or a difference in safety between the combinations.

Because the groups were very small, the counts should be read with particular caution. The registry record does not provide uncertainty estimates or a formal safety comparison. A count of zero serious adverse events in these posted results does not establish that serious events cannot occur, while the reported deaths should not be attributed to a treatment without information on cause and assessment.

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

The posted primary result is straightforward at the level of observed participants: no participant in either arm is recorded as having a complete or partial response under the stated RECIST v1.1 definition. This finding is based on 3 participants assigned to ibrutinib plus cetuximab and 2 assigned to ibrutinib plus nivolumab. It describes the registry results for this trial, not the expected outcome for every person with recurrent or metastatic head and neck squamous cell carcinoma.

The result does not show that the combinations have identical effects. Equal observed response counts in very small groups do not establish equivalence, and the registry record reports no formal treatment contrast or uncertainty interval. Nor does the absence of posted responses establish that responses could not occur in a larger study. The record does not provide enough information to determine the range of population response rates compatible with the data through a reported confidence interval.

The primary endpoint addresses tumor response, while progression-free survival and overall survival are separate outcomes. Although those outcomes are listed as secondary endpoints, their numerical results are not reported in the registry record. The posted response counts therefore cannot answer questions about time to progression or survival. They also do not summarize all possible benefits or harms.

The trial was terminated, but its reason for termination is not reported in the registry record. The posted deaths and adverse-event counts should not be interpreted as proof that treatment caused those outcomes; the record does not provide attribution or timing details. Likewise, the status and small enrollment do not by themselves establish whether the trial ended because of safety, feasibility, recruitment, or another reason.

The most defensible interpretation is limited: the registry reports no qualifying responses among the small number of participants enrolled, alongside incomplete information about analysis methods, follow-up, and event attribution. The results are descriptive and cannot support a definitive ranking of the two combinations.

14. Why This Trial Matters Statistically

This trial illustrates why design labels and observed counts need to be interpreted together. Randomization is an important feature for comparing treatments, but a randomized study with very few participants can still yield highly uncertain results. The arm counts of 3 and 2 make the scale of the available evidence immediately relevant to any interpretation of the response outcome.

The example also highlights the distinction between an endpoint definition and a complete statistical analysis. The primary endpoint is clearly described as best overall response under RECIST v1.1, with stated criteria for complete and partial response. Yet the registry record lists 0 analyses and does not report an effect estimate, confidence interval, or p-value for comparing arms. A defined outcome is necessary for analysis, but it does not alone communicate the uncertainty around a treatment contrast.

Registry completeness matters as well. The record lists two secondary outcomes, but their numerical results are not reported. It does not describe missing-data rules, analysis populations, multiplicity procedures, interim analyses, or the reason for termination. These omissions do not prove that the corresponding procedures were absent, but they limit what can be independently evaluated from the registry record.

Finally, the safety counts show why event counts need context. Serious adverse events, other adverse events, and deaths are reported, but the record does not provide event terms or attribution for deaths. Counts without timing, cause, and an appropriate comparison do not establish a causal relationship. The trial is therefore a useful case for separating what is observed in the registry from what remains uncertain or unreported.

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 most immediate limitation is the very small enrollment: 5 participants in total, with 3 in the ibrutinib-plus-cetuximab arm and 2 in the ibrutinib-plus-nivolumab arm. The posted response count is zero in both groups. Such sparse data provide little information for estimating response in a broader population or for distinguishing the two combinations. The registry record does not report confidence intervals for the primary outcome.

The trial status is terminated, but the reason for termination and the relationship between termination and the planned enrollment or follow-up are not reported in the registry record. This limits assessment of whether the posted results represent complete outcome ascertainment. The planned sample size and sample-size justification are also not reported, so the enrollment cannot be evaluated against a stated target or design assumption.

Important statistical details are absent from the record, including the randomization method, allocation concealment, analysis population, missing-data handling, and the prespecified approach to comparing arms. The record lists 0 analyses. Although progression-free survival and overall survival are named as secondary outcomes, their numerical results are not reported in the registry record.

Outcome measurement details beyond the RECIST v1.1 definitions are not reported. The timing and frequency of imaging, response confirmation, and handling of unevaluable assessments are unknown from the record. Safety reporting is also limited: serious adverse-event terms are not listed, and the causes or treatment relationships of the reported deaths are not given. These limitations restrict both efficacy and safety interpretation and prevent a definitive comparison of the regimens.

16. Related Tutorials

17. Related Statistical Calculators

18. Sources

• ClinicalTrials.gov: Trial of Ibrutinib Combined With Nivolumab or Cetuximab to Treat Recurrent/Metastatic HNSCC — NCT03646461.
• Registry record last updated 2026-05-22; status last verified 2026-04.
• Results first posted 2026-05-22.
• 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

NCT03646461 is a randomized, open-label, phase II parallel-group trial sponsored by the University of California, San Diego. It studied ibrutinib combined with cetuximab or nivolumab in recurrent and/or metastatic head and neck squamous cell carcinoma. The record lists enrollment of 5 and a status of terminated.

The primary endpoint was best overall response under RECIST v1.1, over up to 22 months. The posted response count was zero in each arm: 3 participants in the cetuximab combination arm and 2 in the nivolumab combination arm. Progression-free survival and overall survival were listed as secondary outcomes, but numerical results are not reported in the registry record. The record lists 0 analyses and does not report the reason for termination, a formal between-arm analysis, or key details of missing-data and analysis-population procedures.