Phase II Trial Evaluating the Efficacy of Palbociclib in Combination With Carboplatin for the Treatment of Unresectable Recurrent or Metastatic Head and Neck Squamous Cell Carcinoma: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Phase II Trial Evaluating the Efficacy of Palbociclib in Combination With Carboplatin for the Treatment of Unresectable Recurrent or Metastatic Head and Neck Squamous Cell Carcinoma (NCT03194373), a na phase 2 study. The primary outcome is Percent Disease Control Rate (DCR).

2. Clinical Question

3. Trial Design

This was a multicenter, open-label, single-arm phase II trial sponsored by University of Michigan Rogel Cancer Center. The registry lists the study as completed, with enrollment of 21. The recorded start date is 2017-10-12 and the primary completion date is 2019-02-01. These registry entries describe the study's design and conduct; they do not by themselves establish the outcome of treatment.

All participants were assigned to the palbociclib-and-carboplatin regimen. The initial combination was planned for up to 6 cycles, each lasting 21 days. Palbociclib was listed as 125 mg by mouth daily on days 1-14, while carboplatin was listed as AUC 5 intravenously on day 1. The arm description also mentions maintenance palbociclib after 6 cycles, with 125 mg by mouth daily on days 1-21; the complete maintenance schedule is not reported in the registry record.

Open-label means masking was not used, and single-arm means the study did not include a concurrent control arm. Therefore, the design can describe outcomes observed among treated participants, but it cannot separate treatment effects from the natural course of disease, other care, or participant selection as a randomized comparison could. The registry record does not report a separate control condition, an allocation procedure between groups, or a protocol-based reason for choosing the enrollment target.

Secondary outcomes listed include median progression-free survival time, median overall survival time, and the number of treatment-related toxicities. The registry record does not report the corresponding numerical outcome results in the information available for this analysis. Nor does the record describe enough detail to determine how survival outcomes were analyzed or how treatment-related toxicities were attributed.

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: One enrolled patient withdrew prior to starting study treatment.

Overall Study

Analysis population for the primary outcome, as defined in the registry: Patients were considered evaluable for response if they underwent response evaluation imaging after 2 cycles of receiving both carboplatin and palbociclib.

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: 21 patients enrolled; 18 were evaluable for response. One patient was not started on therapy after enrollment due to abrupt decline in performance status, another received carboplatin but never received palbociclib. A third patient was removed from study prior to imaging per physician discretion.

6. Primary Endpoint

The registry lists the primary outcome as Percent Disease Control Rate (DCR), measured over 12 weeks.

The primary endpoint is the percentage of participants with disease control at 12 weeks. Disease control is a combined response category: complete response, partial response, or stable disease. Complete response requires disappearance of all target lesions with two observations at least 4 weeks apart and no new lesions. Partial response requires at least a 30% decrease in the sum of the longest diameters of target lesions relative to baseline, with no new lesions. Stable disease means the findings meet neither the response criteria nor the progression criteria, using the smallest sum of longest diameters since treatment began as the reference.

A disease control rate combines participants whose disease shrinks with those whose disease does not meet criteria for progression at the assessment time. It therefore answers a broader question than a response rate limited to complete and partial responses. It is used to describe disease status at a specified time point, rather than to measure how long control lasts.

For this endpoint, response evaluability was tied to imaging after 2 cycles of both study drugs. That definition is important when interpreting the denominator: the registry record does not report the number of participants who met it or how participants without the required imaging were handled. The primary endpoint's observed percentage and 95% confidence interval are not reported in the registry record, so the magnitude and precision of the result cannot be assessed from the endpoint definition alone.

7. Why No Re-analysis Is Shown

8. Statistical Interpretation

9. Statistical Methods Explained

Clinical trial results answer different statistical questions depending on the outcome type. The primary outcome in this study is a binary response classification at a fixed assessment time: each evaluable participant either meets the definition of disease control or does not. The disease control rate is the proportion in the disease-control category. It is not a measure of the duration of control, and it does not distinguish the three included response categories unless those categories are reported separately.

What does a confidence interval communicate?

A confidence interval expresses uncertainty around an estimated quantity. For a response proportion, the estimate is based on the observed classifications, while the interval reflects how much uncertainty remains in estimating the underlying rate. A confidence interval is not a guarantee that the true rate lies within its bounds, nor does it show whether an outcome is clinically important. Its width is affected by the amount of usable outcome information and the observed pattern of responses. The registry record specifies a 95% confidence interval for the primary endpoint, but its endpoints and the corresponding estimate are not reported.

Why does the evaluable population matter?

The primary outcome definition says that participants were evaluable if they had response-evaluation imaging after 2 cycles of both drugs. The denominator for a rate should match the population used in the analysis. If participants without the required imaging are excluded, the estimate describes the evaluable group rather than necessarily every enrolled participant. If they are handled by another rule, that could change the estimate. The registry record does not report the number evaluated, the number without assessment, or the rule for handling missing response data. Those omissions make it impossible to reconstruct the analysis population or assess how missing imaging may have affected the result.

What can a single-arm study estimate?

A single-arm trial can estimate the observed outcome distribution among participants receiving the study regimen. It lacks a concurrent group receiving a different treatment or control condition. Thus, even if the observed disease control rate were available, a comparison with an external expectation would require assumptions about whether the populations, assessment methods, eligibility criteria, and care settings were sufficiently comparable. Those assumptions are not detailed in the registry record. The trial's open-label design also means masking was not used, which can matter for outcomes involving judgment in assessment or event attribution.

How should survival outcomes be understood?

Progression-free survival and overall survival are time-to-event outcomes, unlike the fixed-time disease control rate. Their analysis generally considers both whether an event occurred and how long participants were followed; participants without an observed event by the end of follow-up may have incomplete event-time information. General statistical methods for such outcomes account for censoring, but the particular methods, follow-up summary, event counts, and survival estimates for this trial are not reported in the registry record. No survival conclusions can be derived from the outcome names alone.

Why do safety counts need context?

Adverse-event summaries can count participants who experienced an event, the number of events, or the most serious event per participant. These are not interchangeable measures. A participant may experience more than one event, and event severity, timing, and attribution are distinct considerations. The registry safety listing gives participants at risk and affected for serious and other adverse events, along with selected serious-event terms and percentages. It does not explain the detailed event-counting rules, exposure time, or adjudication procedures. In general, longer exposure can provide more opportunity to observe events; treatment duration and exposure summaries are not reported in the registry record.

What is the role of planning illustrations?

Sample-size illustrations are planning calculations, not outcomes observed in the trial. The assumptions stated for the illustrations include a two-sided significance level of 0.05, equal allocation between two arms, and a normal approximation; the effect to detect is described as a fraction of the observed effect. For time-to-event outcomes, the illustrations are described in terms of total events, while other outcomes are expressed as participants per arm. These assumptions do not change the fact that this trial was single-arm. The registry record does not report the numerical planning illustrations or a trial-specific sample-size justification. Any planning estimate should not be read as observed power or as evidence that the trial achieved a particular level of precision.

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("Palbociclib and Carboplatin")
n      <- c(18)
events <- c(6)
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  Palbociclib and Carboplatin  18  6  33.3  13.3  59.0  16.3  56.3  16.3  56.6

11. Multiplicity, Missing Data and Analysis Populations

The registry record lists one primary endpoint and secondary outcomes that include progression-free survival, overall survival, and treatment-related toxicities. When a study examines multiple outcomes or performs multiple statistical tests, the chance of at least one apparently notable result can increase. Statistical methods may address this issue through a prespecified hierarchy, adjusted significance thresholds, or other strategies. The registry record does not report a multiplicity procedure, a testing hierarchy, or whether secondary analyses were considered exploratory. Consequently, the relationship among the outcome analyses cannot be determined from the available record.

Missing outcome data are especially relevant to the primary endpoint because response evaluation required imaging after 2 cycles of receiving both drugs. The record does not report how many participants completed that assessment, how many lacked imaging, why any assessment was unavailable, or how missing responses were treated. Excluding participants without assessment and treating them as not having disease control are different analysis choices; the record does not state which approach was used.

The enrollment listing is 21, while the safety listing records 19 participants at risk for serious and other adverse events. Those are separate reported figures, and the record does not explain the relationship between them. The number included in the primary endpoint analysis is not reported. No analysis-population definitions, interim-analysis details, or missing-data sensitivity analyses are reported in the registry record. These gaps constrain interpretation of both the efficacy and safety summaries.

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: neoplasms benign, malignant and unspecified (incl cysts and polyps) - other (15.8%), dyspnea (10.5%), lung infection (10.5%), nausea (10.5%), pleural effusion (10.5%).

The safety listing for the palbociclib-and-carboplatin group records 19 participants at risk for serious adverse events and 14 affected. It also records 19 participants at risk for other adverse events and 19 affected. These are participant counts, not a measure of how many separate events occurred. The registry record reports 16 deaths, but does not provide the timing, causes, follow-up context, or treatment attribution for those deaths.

Selected serious-event terms are listed with percentages: neoplasms benign, malignant and unspecified (incl cysts and polyps) - other, 15.8%; dyspnea, 10.5%; lung infection, 10.5%; nausea, 10.5%; pleural effusion, 10.5%; and vomiting, 10.5%. These entries describe the registry's reported terms and percentages. The record does not give enough detail to determine whether a participant could appear under multiple terms, how terms were coded, or the time period used for each listing.

Seriousness and treatment relatedness are not synonymous. The listing identifies serious adverse events, but the available information does not establish that each event was caused by treatment. The secondary outcome mentions treatment-related toxicities, yet its numerical results and attribution method are not reported in the registry record. Without exposure duration, event timing, and a comparator, these safety figures cannot provide a comparative risk estimate.

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

The study was designed to estimate disease control at 12 weeks in patients receiving palbociclib with carboplatin. The endpoint combines complete response, partial response, and stable disease, with response assessed using the criteria stated in the registry record. This defines what the trial intended to measure, but the observed disease control rate and 95% confidence interval are not reported in the registry record. The principal efficacy result therefore cannot be quantified or judged for precision from the information available.

Even when an outcome estimate is available, a single-arm study gives a description of the outcomes in its treated group rather than a randomized estimate of the difference between treatments. It cannot rule out explanations related to patient selection, disease course, assessment, or other care. The open-label design is another feature to consider, particularly for outcomes that require interpretation or attribution.

The listed secondary outcomes point to additional questions about time to progression, survival, and toxicities. Their numerical results and analysis details are not reported in the registry record. The safety counts and selected event percentages provide some descriptive information, but they do not establish comparative safety or prove that events were caused by the regimen. The death count alone is not a survival analysis.

Accordingly, the registry record supports understanding the study's aim, regimen, design, endpoint definition, and portions of safety reporting. It does not support a conclusion about the magnitude of disease control, the precision of an efficacy estimate, or a causal effect relative to another treatment. Those questions require outcome results and analysis details that are not reported in the registry record.

14. Why This Trial Matters Statistically

This trial raises a useful statistical distinction between defining an outcome and reporting its estimate. The registry record carefully specifies what qualifies as disease control at 12 weeks, including the response categories and assessment criteria. But an endpoint definition alone cannot show the observed rate, the size of its uncertainty, or how many participants contributed to the estimate. Reporting the analysis denominator and confidence interval is essential for interpreting a proportion.

The study also illustrates the limits of inference from a single-arm design. A response rate can describe the treated group, but there is no randomized comparator to distinguish the regimen's association with outcomes from other influences. This design feature is not repaired by a precise estimate: precision and causal interpretation are different statistical properties. A narrow confidence interval, if one were reported, would still not create a control comparison.

Eligibility for the response analysis depended on imaging after 2 cycles of both drugs. That makes the handling of participants without the required assessment an important part of interpretation. Without the denominator and missing-data rules, readers cannot determine precisely which population the primary rate represents. This is a general lesson for fixed-time outcomes: assessment schedules and analysis-population definitions belong alongside the headline estimate.

The safety information further shows why counts need definitions and context. Counts of affected participants, serious-event terms, deaths, and treatment-related toxicities answer different questions. Without further detail, they should remain descriptive rather than being treated as comparative or causal evidence. The registry record's design and reporting features therefore help identify what can be learned and what remains uncertain.

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 key limitation for interpreting efficacy is that the registry record defines the primary endpoint but does not report the observed disease control rate or its 95% confidence interval. It also does not report the number of participants included in the endpoint analysis. Although enrollment is listed as 21, response evaluability required imaging after 2 cycles of both drugs, and the number meeting that definition is not reported.

The trial had no concurrent control group and was open label. As a result, any observed outcome would be descriptive of the treated study group and could not by itself establish a causal difference from another treatment or control condition. The registry record does not report details of blinded response review or other procedures that might address assessment concerns.

The handling of missing response assessments, reasons for missingness, and analysis-population rules is not reported in the registry record. Numerical results for the listed survival outcomes and treatment-related toxicities are also not reported. The record does not give survival follow-up details, event definitions for those analyses, or the method used to attribute toxicities to treatment.

Safety reporting includes participant counts, selected serious-event terms, percentages, and a death count, but lacks the context needed to assess timing, duration of exposure, causes, or comparative risk. The full maintenance schedule is not reported in the registry record. A trial-specific sample-size justification, multiplicity approach, and interim-analysis details are also not reported. These limitations restrict both reconstruction of the analyses and interpretation of the reported study design.

16. Related Tutorials

17. Related Statistical Calculators

18. Sources

• ClinicalTrials.gov: Phase II Trial Evaluating the Efficacy of Palbociclib in Combination With Carboplatin for the Treatment of Unresectable Recurrent or Metastatic Head and Neck Squamous Cell Carcinoma — NCT03194373.
• Registry record last updated 2021-04-12; status last verified 2021-04.
• Results first posted 2020-02-05.
• 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, multicenter, open-label, single-arm phase II study of palbociclib with carboplatin for unresectable recurrent or metastatic head and neck squamous cell carcinoma. Enrollment is listed as 21. The primary objective was to estimate disease control at 12 weeks, counting complete response, partial response, or stable disease. Response evaluability required imaging after 2 cycles of both drugs.

The registry record does not report the observed primary endpoint estimate or its 95% confidence interval. It also does not report numerical results for the listed survival outcomes. Safety listings include serious and other adverse-event counts, selected serious-event terms and percentages, and 16 deaths, without sufficient detail to establish comparative risk or treatment causality. The single-arm design supports descriptive interpretation of the treated group, not a randomized treatment comparison.