Trial of Nivolumab as a Novel Neoadjuvant Pre-Surgical Therapy for Locally Advanced Oral Cavity Cancer: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Trial of Nivolumab as a Novel Neoadjuvant Pre-Surgical Therapy for Locally Advanced Oral Cavity Cancer (NCT03021993), a na phase 2 study. The primary outcome is Objective Response Rate Using Pathological Response.

2. Clinical Question

The trial asks whether nivolumab given before surgery is associated with evidence of tumor response in the surgical specimen. This is a question about pathological findings after treatment, not a randomized comparison of nivolumab with another treatment or with no pre-surgical treatment. The registry describes the study as phase II and gives an enrollment of 17.

The endpoint combines two categories. A complete pathological response requires no invasive and no in situ residuals in the specimen. A partial pathological response requires at least a 30% reduction in lesion size, determined by comparing pretreatment clinical measurements with final pathological measurements. The primary outcome is measured at surgery, with the timeframe stated as day 36 or day 50.

The registry record does not report how many participants met either response criterion, a response proportion, or an uncertainty interval for the response result. It therefore does not answer the primary question quantitatively in the posted results. The trial’s design can identify a response pattern within its treated group, but without a control group it cannot isolate the effect of nivolumab from other explanations for observed pathological findings.

3. Trial Design

This is a phase II, completed, single-group trial of nivolumab as neoadjuvant, pre-surgical treatment for locally advanced oral cavity cancer. The registry records an enrollment of 17 and identifies Oral Cavity SCC as the condition. The sponsor is the Medical University of South Carolina. The recorded study start is 2017-05-30, and primary completion is recorded as 2021-11-15.

Participants in the nivolumab group were to receive 3mg/kg every two weeks for up to four doses before surgery. The registry’s primary endpoint is assessed at surgery, with a stated timeframe of day 36 or day 50. The record describes no masking and no randomized allocation. With a single group and no comparator, the design is suited to describing outcomes among people who received the pre-surgical regimen, not to making a like-for-like estimate against an untreated or alternative-treatment group.

The main endpoint is an objective response rate based on pathological response. The definition combines complete response and partial response, with the latter requiring at least a 30% reduction between pretreatment clinical measurement and final pathological measurement. The registry also lists biological secondary outcomes concerning peripheral-blood immune cells and cytokines, including Treg cells, activated T-cells, IFN-gamma, and Granzyme-B expression.

The record marks the study completed, but completion status alone does not establish that every planned measurement was obtained or that all results were posted. The registry record does not report a detailed statistical analysis plan, the analysis population for the primary endpoint, or the number of participants with an evaluable surgical specimen. It also does not report a control-group design or a treatment allocation procedure beyond the single-group designation.

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

Reasons for not completing this period:

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 Objective Response Rate Using Pathological Response, measured over Time of surgery (day 36 or day 50).

The primary endpoint is the count of participants with an objective pathological response at surgery. The response definition includes either a complete pathological response, with no invasive and no in situ residuals in the surgical specimen, or a partial pathological response, defined as at least a 30% reduction in lesion size. That reduction is assessed by comparing pretreatment clinical measurements with final pathological measurements.

Pathological response is used because the surgical specimen allows assessment of residual disease after the pre-surgical treatment. This gives a tissue-based measure of response at a clinically defined point in the treatment pathway. Combining complete and partial responses creates a single endpoint intended to capture more than complete eradication of detectable residual disease while retaining a stated threshold for partial response.

As a general statistical matter, a count endpoint records how many participants meet a defined event criterion. A response rate would express that count relative to an identified analysis group, and an interval estimate would convey uncertainty around the rate. The registry record identifies the outcome as a participant count but does not report the observed response count, the denominator used for a response rate, or an interval estimate.

The endpoint’s interpretation also depends on how the clinical and pathological measurements were obtained and matched. Details about measurement procedures, handling of missing specimens, and adjudication of response are not reported in the registry record. The stated timeframe is time of surgery (day 36 or day 50); the record does not explain how those alternative timepoints relate to individual treatment schedules.

7. Why No Re-analysis Is Shown

8. Statistical Interpretation

9. Statistical Methods Explained

The registry identifies the primary measure as a count of participants and classifies the outcome as binary. In general, a binary outcome assigns each participant to one of two states for the endpoint: the defined event occurred, or it did not. For this trial, the event is meeting either of the stated pathological response criteria at surgery. The record does not post the event count or identify the analysis denominator, so a response proportion cannot be interpreted from the information available.

What would a response estimate represent?

In general, a response proportion is the number of participants meeting the response definition divided by the number included in the endpoint analysis. The denominator matters: it could be defined by enrollment, treatment received, surgery, or availability of an evaluable specimen, depending on the analysis plan. These choices can lead to different interpretations if some participants lack an endpoint measurement or do not undergo the planned assessment. The registry record does not report the primary endpoint denominator or the analysis population, so the basis for such a proportion is not specified.

An uncertainty interval is ordinarily reported alongside a proportion. In general, a confidence interval expresses the precision of an estimate under a statistical model and method; a wider interval indicates less precision. It does not guarantee that the true value lies within the interval, nor does it address biases that arise from study design or measurement. No confidence interval for pathological response is reported in the registry record. Without the result and interval, neither the observed response level nor its statistical precision can be assessed.

Why does the single-group design matter?

A single-group study estimates outcomes within the treated group. In a randomized comparison, treatment groups can provide a basis for estimating a contrast, subject to the study’s design and analysis. This trial is not recorded as randomized and has no control arm. Therefore, an outcome observed after nivolumab cannot be attributed to nivolumab by comparison with a concurrent group. General factors such as disease characteristics, selection into the study, measurement variation, and other care may also be relevant to interpretation. The registry record does not provide data that would quantify or resolve these possibilities.

Masking is recorded as none. In general, masking can reduce the influence of knowledge about treatment on some assessments or decisions, depending on which people are masked and how outcomes are measured. Pathological assessment may involve judgment in classifying residual disease, but the record does not report assessor procedures or whether response assessments were independently reviewed. No further conclusion about assessment bias can be drawn from the masking field alone.

How should the response definition be understood?

The endpoint combines complete and partial pathological response. The complete-response criterion is absence of invasive and in situ residuals in the surgical specimen. The partial-response criterion uses at least a 30% reduction in lesion size, comparing pretreatment clinical measurements with final pathological measurements. A combined outcome can summarize more than one response state, but its meaning depends on the definition and on consistent measurement. The registry record does not report separate counts for complete and partial response, measurement reliability, or how borderline cases were classified.

The timing is specified as time of surgery (day 36 or day 50). In general, a defined assessment window helps establish when an outcome is measured relative to treatment. If assessment timing varies, the interpretation may depend on the actual timing and the rules for assigning an endpoint. The registry record does not explain the distinction between the two stated days or report individual assessment times.

What do the planning calculations mean?

General sample-size planning uses assumptions about the outcome, target effect, significance level, allocation, and statistical method to estimate how many participants a future study might need. The record describes its planning illustration as based on the observed effect, with a two-sided significance level of 0.05, equal allocation between two arms, and a normal approximation; the effect to detect is a stated fraction of the observed effect. These are planning assumptions, not evidence that the trial had a particular probability of detecting an effect after it was conducted. Sample-size illustrations are not observed power.

The stated equal allocation between two arms does not match the registered single-group design. The illustration should therefore be read as a hypothetical planning scenario rather than as a description of the trial’s actual allocation or analysis. The record says that sample-size outputs for non-time-to-event outcomes are participants per arm, while time-to-event outputs are total events; the primary endpoint is a participant count, not a time-to-event outcome. The trial’s own sample-size justification is not reported in the registry record.

More broadly, a statistical analysis plan would specify the endpoint analysis, analysis population, missing-data handling, and any adjustment for multiple outcomes or interim looks. General statistical practice relies on these details to understand which observations contribute and how results are interpreted. Such a plan, any interim analyses, and a missing-data strategy are not reported in the registry record. The posted information is therefore insufficient to reconstruct a primary endpoint analysis or judge whether it followed a prespecified plan.

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("Nivolumab")
n      <- c(12)
events <- c(4)
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  Nivolumab  12  4  33.3  9.9  65.1  13.8  60.9  13.9  61.4

11. Multiplicity, Missing Data and Analysis Populations

The registry lists 7 outcome measures. The primary endpoint is pathological response, and the listed secondary measures concern immune-cell levels and immune markers in peripheral blood. When a study assesses multiple outcomes, general statistical practice considers multiplicity: testing several hypotheses can increase the chance of at least one apparently notable result arising by chance. A plan may specify which outcomes are primary, whether secondary results are exploratory, and whether any adjustment is used. Multiplicity handling is not reported in the registry record.

The record lists 0 analyses. This means analysis details are not present in the registry record; it does not establish whether analyses were conducted elsewhere. No primary endpoint model, confidence-interval method, hypothesis-testing procedure, or analysis population is reported. The record also does not report interim analyses or stopping rules.

Missing-data handling is not described. In general, a missing endpoint can affect the interpretation of a response proportion because the denominator and assumptions about unobserved outcomes matter. The registry record does not say how participants without an evaluable surgical specimen, incomplete measurements, or other unavailable outcomes were handled. It also does not define an intention-to-treat, per-protocol, or other analysis population. These omissions prevent assessment of whether missing observations could have influenced the result, and no missingness pattern can be inferred from the posted information.

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: hypercalcemia (5.9%).

The registry lists safety counts for the nivolumab group. For serious adverse events, 1 participant was affected among 17 at risk. For other adverse events, 17 participants were affected among 17 at risk. The record also lists 3 deaths. Hypercalcemia appears among the top serious terms at 5.9%.

These are descriptive registry figures. They do not by themselves indicate that nivolumab caused an event or death. The record does not provide the timing, event-level attribution, severity details beyond the serious-event classification, or a comparator group for these counts. The listed deaths should not be interpreted as treatment-related on the basis of this information alone.

In general, safety counts are interpreted in context of how events were defined, collected, and attributed, as well as the period of observation and the population at risk. The registry record does not report those details sufficiently to make comparative or causal conclusions. The hypercalcemia percentage is reported as 5.9%; it should be read as a registry-listed serious term, not as a standalone estimate of treatment-attributable risk.

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

The record establishes that a completed phase II, single-group study evaluated nivolumab before surgery for locally advanced oral cavity cancer. It describes the planned regimen, the pathological response definition, the assessment timeframe, biological secondary outcomes, and selected safety counts. It does not report the primary response count, response rate, or uncertainty interval. Consequently, the magnitude and precision of pathological response cannot be assessed from the posted results.

Even if a response result were available, the single-group design would support a description of outcomes among treated participants rather than a randomized estimate of what nivolumab changed. Without a concurrent comparator, the observed outcome cannot separate treatment effects from other influences. The absence of masking may also be relevant to assessment, but the registry does not report enough about pathological review to determine its practical impact.

The safety figures are counts recorded for the nivolumab group, including serious and other adverse events and deaths. They do not establish causation or comparative safety. No control group is available, and the registry record does not provide the event timing, attribution, or a detailed safety analysis.

The study should therefore be read as a record of a pre-surgical, single-group investigation with a defined tissue-based endpoint, not as a definitive comparative assessment. The registry record does not report a statistical analysis plan, handling of missing endpoints, multiplicity procedures, or the analysis population. Those gaps limit evaluation of the reliability and scope of any endpoint conclusions. No treatment recommendation or clinical conclusion follows from the design and posted figures alone.

14. Why This Trial Matters Statistically

This trial is statistically notable because its primary outcome is a pathological response measured after pre-surgical treatment, while the design has no control group. That combination makes the distinction between description and causal comparison central. A single-group result can characterize responses in the studied group, but it does not estimate a contrast against an alternative treatment or untreated course.

The endpoint also combines complete and partial pathological response. The criteria are explicit in the registry record, including the requirement for no invasive and no in situ residuals for complete response and at least a 30% reduction for partial response. Such definitions are essential to make an endpoint interpretable. At the same time, the combined measure conceals the distribution between response categories unless those results are reported separately; separate counts are not reported in the registry record.

Precision is as important as the point estimate in a phase II study. A response count without a denominator, or a proportion without an interval, would not show how uncertain the estimate is. In this record, the response result and interval are not reported, so that statistical assessment is not possible.

The record also lists immune-related secondary outcomes and safety information, illustrating that trial interpretation extends beyond the primary endpoint. Their meaning depends on outcome definitions, analysis methods, missing-data handling, and multiplicity considerations. Those statistical details are not reported. The case therefore highlights how a clear endpoint definition and trial status do not, on their own, supply the numerical results and analysis context needed to judge evidence.

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 limitation for interpreting efficacy is that the registry record does not report the pathological response result. The response count, denominator, response rate, and uncertainty interval are unavailable. The analysis population and handling of participants without an evaluable surgical specimen are also not reported, so the completeness of endpoint ascertainment cannot be assessed.

The single-group design has no randomized or concurrent comparator. Any observed outcomes could not be contrasted with outcomes under another treatment using this study design. The record describes no masking; it does not explain how pathological response was assessed, whether reviewers were independent, or how disagreements were resolved.

The primary endpoint compares pretreatment clinical measurements with final pathological measurements. Measurement procedures and the handling of any differences between those assessments are not detailed in the registry record. The timeframe is stated as day 36 or day 50, but the record does not explain the reason for the alternative timing or report participant-level timing.

Statistical limitations include the absence of a reported analysis plan, interim-analysis information, multiplicity procedures, missing-data methods, and analysis-population definition. The listed secondary outcomes are not accompanied by results or analysis details. The planning illustration uses equal allocation between two arms, whereas the registered trial is single-group; it should not be mistaken for the trial’s own design or sample-size justification. The study’s sample-size rationale is not reported in the registry record.

Safety interpretation is constrained by the lack of a comparator and limited event context. The record does not provide enough information on event timing, attribution, or ascertainment to support causal conclusions from the listed counts.

16. Related Tutorials

17. Related Statistical Calculators

18. Sources

• ClinicalTrials.gov: Trial of Nivolumab as a Novel Neoadjuvant Pre-Surgical Therapy for Locally Advanced Oral Cavity Cancer — NCT03021993.
• Registry record last updated 2025-11-12; status last verified 2025-10.
• Results first posted 2025-11-12.
• 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

NCT03021993 is a completed phase II, single-group trial of nivolumab before surgery for locally advanced oral cavity cancer. The registry records an enrollment of 17, no masking, and no randomized allocation. The regimen is described as 3mg/kg every two weeks for up to four pre-surgical doses.

The primary outcome is pathological response at surgery (day 36 or day 50), defined as complete response or partial response meeting the stated 30% reduction criterion. The registry record does not report the primary response result, its denominator, or an uncertainty interval. It lists 7 outcome measures and 0 analyses.

Safety counts include 1 participant with a serious adverse event among 17 at risk, 17 with other adverse events among 17 at risk, and 3 deaths. Hypercalcemia is listed at 5.9% among the top serious terms. These figures are descriptive and do not establish causation. The record does not report a control-group comparison, the trial’s statistical analysis plan, missing-data methods, or a detailed sample-size justification.