This page provides an independent statistical analysis and educational interpretation of publicly reported results. ClinicalTrials.gov provides the official trial registry record. Numerical trial results on this page are restricted to the results posted in the ClinicalTrials.gov record; calculations added by Clinical Biostats are labelled as such.
1. Trial at a Glance
This page covers the registry results for Pembrolizumab in Treating Patients With High Risk Oral Intraepithelial Neoplasia (NCT02882282), a randomized phase 2 study. The primary outcome is Oral Cancer-free Survival.
| Feature | Registry value |
|---|---|
| ClinicalTrials.gov identifier | NCT02882282 |
| Phase | Phase 2 |
| Condition | Oral Cavity Carcinoma, Oral Intraepithelial Neoplasia |
| Allocation | Randomized |
| Design model | Parallel |
| Masking | None (open label) |
| Enrollment | 15 |
| Arms in the primary outcome | 2 |
| Lead sponsor | M.D. Anderson Cancer Center |
| Results posted | Yes |
2. Clinical Question
The study asks whether oral cancer-free survival and overall survival differ between patients with high-risk oral intraepithelial neoplasia assigned to pembrolizumab and patients assigned to observation. It is a randomized phase 2 trial, but the registry record does not report an analysis comparing the two groups. The posted primary endpoint result is a median oral cancer-free survival estimate for the pembrolizumab arm; the observation-arm value is not reported as an estimable result.
Population
Patients with high-risk oral intraepithelial neoplasia. The primary endpoint population is described as intent-to-treat. The record lists an enrollment of 15.
Intervention
Pembrolizumab, administered intravenously over 30 minutes on day 1 and repeated every 21 days for up to 4 courses, unless disease progression or unacceptable toxicity occurs.
Comparator
Observation. The registry describes this group as undergoing observation; it does not give further details about observation procedures.
Primary question
What is oral cancer-free survival for patients assigned to pembrolizumab versus observation? Overall survival is listed as a secondary outcome, but a secondary outcome result is not posted.
The question is clinically focused but the available results do not establish a comparative effect. Without an observation-arm estimate or a reported between-group analysis, the pembrolizumab median cannot be interpreted as evidence of a difference from observation. The registry record also does not describe how oral cancer-free survival events were defined, how follow-up was handled, or what criteria determined the timing of an event.
3. Trial Design
This is a randomized, parallel-group phase 2 study sponsored by M.D. Anderson Cancer Center. The registry lists enrollment of 15 and reports the study as completed. The start date is 2017-06-14 and the primary completion date is 2025-07-14. The study has no masking listed, so the design is unmasked in the registry record. Participants were assigned to observation or pembrolizumab.
In the observation arm, patients undergo observation. In the pembrolizumab arm, patients receive the drug intravenously over 30 minutes on day 1, with treatment repeating every 21 days for up to 4 courses in the absence of disease progression or unacceptable toxicity. The record does not describe allocation ratios, the randomization procedure, concealment of assignments, or additional features of observation. It also does not explain the personalized aspect referenced in the official title.
The primary outcome is oral cancer-free survival, measured in months, with a stated timeframe of 51.2 months. The primary endpoint population is described as intent-to-treat. Overall survival is listed as a secondary outcome. Although the record lists two outcome measures, it posts only one secondary outcome and provides no analysis results for it. The registry record does not report a protocol-level statistical analysis plan, a target enrollment justification, or any interim-analysis plan. The enrollment figure describes the reported study enrollment; it should not be taken as evidence about how the trial’s planned sample size was chosen.
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
| Stage | Arm A (Control) | Arm B (Pembrolizumab). |
|---|---|---|
| STARTED | 7 | 8 |
| COMPLETED | 7 | 7 |
| NOT COMPLETED | 0 | 1 |
Reasons for not completing this period:
| Reason | Arm A (Control) | Arm B (Pembrolizumab). |
|---|---|---|
| Adverse Event | 1 | 0 |
Analysis population for the primary outcome, as defined in the registry: Intent-to-treat population
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.
| Characteristic | Unit | Arm A (Control) | Arm B (Pembrolizumab) | Total |
|---|---|---|---|---|
| Participants analysed | 7 | 8 | 15 | |
| Age, Categorical: <=18 years | Participants | 0 | 0 | 0 |
| Age, Categorical: Between 18 and 65 years | Participants | 4 | 4 | 8 |
| Age, Categorical: >=65 years | Participants | 3 | 4 | 7 |
| Sex: Female, Male: Female | Participants | 2 | 7 | 9 |
| Sex: Female, Male: Male | Participants | 5 | 1 | 6 |
| Ethnicity (NIH/OMB): Hispanic or Latino | Participants | 0 | 0 | 0 |
| Ethnicity (NIH/OMB): Not Hispanic or Latino | Participants | 7 | 7 | 14 |
| Ethnicity (NIH/OMB): Unknown or Not Reported | Participants | 0 | 1 | 1 |
| Race (NIH/OMB): American Indian or Alaska Native | Participants | 0 | 0 | 0 |
| Race (NIH/OMB): Asian | Participants | 0 | 1 | 1 |
| Race (NIH/OMB): Native Hawaiian or Other Pacific Islander | Participants | 0 | 0 | 0 |
| Race (NIH/OMB): Black or African American | Participants | 0 | 0 | 0 |
| Race (NIH/OMB): White | Participants | 7 | 7 | 14 |
| Race (NIH/OMB): More than one race | Participants | 0 | 0 | 0 |
| Race (NIH/OMB): Unknown or Not Reported | Participants | 0 | 0 | 0 |
| Region of Enrollment: United States | participants | 7 | 8 | 15 |
6. Primary Endpoint
The registry lists the primary outcome as Oral Cancer-free Survival, measured over 51.2 months.
| Element | Registry entry |
|---|---|
| Result type | Median |
| Measure of spread | 95% Confidence Interval |
| Unit | Months |
| Endpoint classification | Time-to-Event Analysis |
Oral cancer-free survival is the primary endpoint. It is a time-to-event measure intended to describe the time until an event that qualifies as oral cancer occurs, with the result expressed in months. The registry outcome description also refers to overall survival, but that is listed separately as a secondary outcome. The record does not define the qualifying oral cancer event or state whether particular diagnoses, recurrences, or other events count toward the endpoint.
A time-to-event endpoint can account for both whether an event was observed and how long a participant was followed before an event or the end of observation. This is useful when participants have different follow-up durations. In general, a median time-to-event estimate is the point at which the estimated event-free proportion reaches the halfway point, under the analysis method used. A confidence interval describes uncertainty around an estimate; it does not show the range of individual participants’ times.
The registry reports a median oral cancer-free survival of 6.7 months for the pembrolizumab arm, with a 95% confidence interval from 3.6 to 40.2 months. It does not report an estimable median for observation. Because the record does not provide an endpoint definition, censoring rules, or a between-arm analysis, the estimate alone cannot establish how oral cancer-free survival compares between the randomized groups.
7. Why No Re-analysis Is Shown
8. Statistical Interpretation
The posted primary endpoint result is a median oral cancer-free survival of 6.7 months for the pembrolizumab arm, with a 95% confidence interval from 3.6 to 40.2 months. The interval is broad, indicating substantial uncertainty about the median based on the posted estimate. It is an interval for the arm-specific median, not a range containing most participants’ survival times and not an estimate of the difference between treatment groups.
The registry does not give an estimable median for the observation arm. It also does not report a hazard ratio, a confidence interval for a between-arm effect, or a p-value for the primary endpoint. Therefore, the posted result does not quantify whether the groups differ or how large any difference might be. Comparing the pembrolizumab median with a missing observation result would not be a valid substitute for a randomized comparison.
Enrollment was 15, with 7 participants listed in the observation arm and 8 in the pembrolizumab arm. These group sizes give context for the limited amount of information available, but they do not by themselves explain why the observation result is unavailable. The registry does not report the number of oral cancer events, the number censored, or the follow-up distribution by arm. It is therefore not possible to identify from the record why the median was not estimable for observation.
Overall survival is listed as a secondary outcome, but no secondary outcome result is posted. The record also contains no posted analyses. Thus, the outcome listings should not be mistaken for completed statistical comparisons: an outcome can be registered without a corresponding result being available.
The safety record lists no participants with a serious adverse event in either arm. It lists other adverse events for 0 of 7 participants in the observation arm and 6 of 8 in the pembrolizumab arm. Deaths are reported for 1 participant in the observation arm and 7 in the pembrolizumab arm. The record does not specify causes, timing, or attribution for these deaths, so they should not be interpreted as treatment-related findings. Overall, the registry supports a description of the posted arm-specific estimates and event counts, but not a conclusion about comparative efficacy or causation.
9. Statistical Methods Explained
The registry identifies a randomized, parallel-group design with no masking listed. Randomization is intended to make groups comparable, on average, with respect to both measured and unmeasured characteristics. In a small trial, chance differences between groups can still occur. The record does not describe how randomization was implemented, whether allocation was concealed, or whether any stratification factors were used. Those details are not reported in the registry record.
What does a median time-to-event estimate describe?
General explanation: a time-to-event analysis considers both the occurrence of an event and the time until that event. Participants who have not had an event by the end of their available follow-up may contribute information up to that point without being treated as if an event occurred. This is commonly called censoring. The registry record does not state the censoring rules or the method used to calculate the median. It reports a median oral cancer-free survival of 6.7 months in the pembrolizumab arm, with a 95% confidence interval from 3.6 to 40.2 months. No median is reported as estimable for observation.
A median is a summary of the estimated time-to-event distribution, rather than an average of every participant’s observed time. A confidence interval represents uncertainty around the estimate under the statistical method used. The interval from 3.6 to 40.2 months is wide, so the reported median has limited precision. The interval does not give the times experienced by individual participants, nor does it estimate a treatment-versus-observation difference.
Why is a between-arm estimate important?
General explanation: a randomized trial’s treatment question concerns a contrast between groups. A comparison may be expressed using a difference in event-free time, a hazard ratio, or another prespecified measure. Each measure answers a distinct statistical question and depends on a defined analysis approach. The registry record reports no hazard ratio, p-value, or other between-arm estimate, and it does not provide an estimable observation-arm median. Consequently, no comparative effect can be interpreted from the posted primary endpoint result. The record does not report why the observation-arm median is unavailable.
In general, a hazard ratio compares event rates over follow-up under a model and assumptions; it is not the same as a ratio of medians. The registry record does not provide a hazard ratio for this trial, so there is no sponsor-reported hazard ratio to interpret. It would not be appropriate to infer one from the median in a single arm, and no recalculated comparison is reported.
How should uncertainty and statistical significance be read?
General explanation: a confidence interval conveys the range of estimates compatible with the observed data and the method’s assumptions at the stated confidence level. A narrower interval indicates more precision than a wider interval, all else equal. A p-value, when reported, addresses the compatibility of data with a specified null hypothesis under a statistical model; it is not the probability that a treatment works. Neither a p-value nor a formal between-arm confidence interval is posted for the primary endpoint. The 95% confidence interval for the pembrolizumab-arm median cannot be used to determine whether there is a statistically significant difference from observation.
The record identifies the primary population as intent-to-treat. In general, an intent-to-treat analysis retains participants in their assigned groups, regardless of treatment received, according to the analysis definition. The registry does not state how this population was operationalized, whether all randomized participants were included in the posted endpoint estimate, or how missing outcome data were handled. Those choices can affect a time-to-event analysis and its interpretation.
What do the planning assumptions mean?
The stated planning illustration uses a two-sided significance level of 0.05, equal allocation between groups, and a normal approximation, with the effect to detect expressed as a fraction of the observed effect. For time-to-event outcomes, the illustration is based on the total number of events. These are planning assumptions, not evidence about the trial’s original design process or the precision achieved in the enrolled sample. The registry record does not report a trial-specific sample-size justification. A planning illustration is not observed power and does not change the interpretation of the posted endpoint results.
The registry lists no analyses. As a result, the posted outcome information does not supply the analysis details needed to reproduce a between-group comparison. The endpoint definition, statistical model, censoring rules, treatment of missing data, and adjustment strategy are not reported in the registry record. Without those details and a reported comparative estimate, the most defensible statistical reading is limited to describing the pembrolizumab-arm median and its uncertainty, noting the unavailable observation-arm estimate, and distinguishing both from any claim about treatment effect.
10. Multiplicity, Missing Data and Analysis Populations
The registry lists oral cancer-free survival as the primary outcome and overall survival as a secondary outcome. It reports two outcome measures and one posted secondary outcome, but no analyses are posted. The listing of multiple outcomes raises the general issue of multiplicity: when several outcomes or comparisons are tested, the chance of at least one apparently unusual result can increase. Statistical plans may address this through prespecified testing hierarchies or other adjustments. The registry record does not report whether any multiplicity strategy was planned or used.
The primary population is described as intent-to-treat. The record does not state the number of participants included in the primary endpoint analysis, whether the listed arm sizes correspond to the analysis population, or how participants with missing outcome information were handled. Rules for censoring, loss to follow-up, and incomplete endpoint ascertainment are not reported in the registry record. These details matter for time-to-event outcomes because follow-up information may be incomplete even when a participant has not experienced the event.
The observation-arm median is not reported as estimable. The record does not explain whether this reflects the observed event pattern, follow-up, missing outcome data, or another reason. No sensitivity analyses are posted. It is therefore not possible to assess how alternative missing-data assumptions or analysis choices would affect the endpoint result. Overall survival is listed as secondary, but no result is posted for it; the listing alone does not establish that the outcome was analyzed or that a result was statistically significant.
11. Safety Results
The registry reports serious adverse events as affected participants divided by participants at risk.
| Arm | Participants with serious adverse events | Participants at risk | Percentage |
|---|---|---|---|
| Arm A | 0 | 7 | 0.0% |
| Arm B | 0 | 8 | 0.0% |
Deaths
| Arm | Deaths (participants affected of at risk) |
|---|---|
| Arm A | 1 of 7 |
| Arm B | 7 of 8 |
Other (non-serious) adverse events
Participants with at least one non-serious event at or above the registry's 0% reporting threshold.
| Arm | Participants affected of at risk |
|---|---|
| Arm A | 0 of 7 |
| Arm B | 6 of 8 |
Most frequent non-serious adverse event terms
| Event term | Organ system | Arm A | Arm B |
|---|---|---|---|
| Constipation | Gastrointestinal disorders | 0 of 7 | 4 of 8 |
| Fatigue | General disorders | 0 of 7 | 3 of 8 |
| Mucositis Oral | Gastrointestinal disorders | 0 of 7 | 2 of 8 |
| Hypothyroidism | General disorders | 0 of 7 | 2 of 8 |
| Nausea | Gastrointestinal disorders | 0 of 7 | 2 of 8 |
| Pruritus | Skin and subcutaneous tissue disorders | 0 of 7 | 2 of 8 |
| Non-Cardiac chest pain | Gastrointestinal disorders | 0 of 7 | 1 of 8 |
| Sepsis | Infections and infestations | 0 of 7 | 1 of 8 |
| Allergic Rhinitis | Respiratory, thoracic and mediastinal disorders | 0 of 7 | 1 of 8 |
| Anorexia | Metabolism and nutrition disorders | 0 of 7 | 1 of 8 |
The serious adverse event counts are 0 of 7 participants in the observation arm and 0 of 8 in the pembrolizumab arm. No serious adverse event terms are listed. These figures describe the registry’s recorded counts; they do not establish that no events occurred outside the recorded definitions or reporting period.
For other adverse events, the record lists 0 of 7 participants in the observation arm and 6 of 8 in the pembrolizumab arm. It also reports deaths for 1 participant in the observation arm and 7 in the pembrolizumab arm. The registry does not report the causes or timing of the deaths, or whether they were considered related to treatment. It does not provide event-level descriptions or comparative safety analyses in the posted results.
These are counts from a trial with enrollment of 15. They should be read as descriptive observations, not as precise estimates of risk or evidence of causation. The registry record does not provide enough information to determine why the death counts differ between arms or to attribute any death to pembrolizumab.
12. What the Result Does, and Does Not, Mean
The result establishes that the registry reports a median oral cancer-free survival estimate of 6.7 months for the pembrolizumab arm, with a 95% confidence interval from 3.6 to 40.2 months. It also establishes that the registry does not report an estimable median for the observation arm. The interval conveys considerable uncertainty around the arm-specific median; it does not describe individual participants’ outcomes or the size of a difference between groups.
The result does not establish that pembrolizumab improves, worsens, or has no effect on oral cancer-free survival compared with observation. Such a conclusion would require a reported comparison and its uncertainty, neither of which is posted. The absence of an estimable observation-arm median is not itself evidence that the groups are equal or different. The record does not explain why that estimate is unavailable.
Overall survival appears as a secondary outcome, but no result is posted. The registry also does not provide a hazard ratio or p-value for the primary comparison, a count of oral cancer events, or details of censoring and follow-up. These omissions limit what can be inferred from the reported endpoint.
The safety counts describe recorded participants with serious or other adverse events and deaths. They do not identify causes of death or establish whether events were treatment-related. The counts should not be used to infer a causal safety effect from this record alone.
More generally, a randomized design can support comparative inference when outcomes are analyzed and reported with appropriate methods, but randomization alone does not supply a result that is absent from the registry. The available record supports a cautious description of the reported median and the safety counts. It does not support a treatment recommendation or a definitive comparative conclusion.
13. Why This Trial Matters Statistically
This study is statistically noteworthy because it asks a randomized comparative question but posts only an arm-specific primary endpoint estimate. The contrast of pembrolizumab with observation is central to the study question, yet the observation-arm median is not reported as estimable and no between-arm analysis is posted. This illustrates the distinction between registering an outcome and reporting an interpretable comparison.
The confidence interval around the pembrolizumab-arm median extends from 3.6 to 40.2 months. Its breadth makes precision an important part of the interpretation: a single median should not be read as a sharply determined value. The absence of a corresponding observation result makes it impossible to assess the comparative effect from the posted endpoint data. A confidence interval for one arm cannot substitute for uncertainty around a contrast.
The study also highlights why outcome definitions and analysis details matter. The registry names oral cancer-free survival but does not specify which events qualify, how censoring was handled, or how missing outcome information was treated. Those details affect the meaning of a time-to-event estimate and the ability of readers to evaluate or reproduce an analysis.
Finally, the record lists overall survival as a secondary outcome without posting its result. Recording multiple outcomes does not show that each has been analyzed or reported. Clear outcome reporting, analysis methods, and uncertainty estimates are essential for distinguishing a descriptive finding from evidence about a treatment comparison. The posted record allows the former, but not the latter, for the primary question.
14. 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 main limitation is that the registry reports no estimable oral cancer-free survival median for observation and no analysis comparing the randomized groups. The available primary endpoint estimate is therefore restricted to the pembrolizumab arm. No hazard ratio, p-value, or confidence interval for a between-arm effect is reported. The record does not explain why the observation estimate is unavailable.
The endpoint description does not define the events counted as oral cancer-free survival events. The record also does not report censoring rules, follow-up distributions by arm, event counts, or the handling of missing outcome information. Without these details, readers cannot assess how the time-to-event estimate was constructed or how robust it may be to analysis choices.
Overall survival is listed as a secondary outcome, but no result is posted. No analyses are posted, and the record does not describe a multiplicity approach or sensitivity analyses. The primary population is called intent-to-treat, but the record does not give further detail about the analyzed population or exclusions.
The trial enrolled 15 participants, with 7 listed in observation and 8 in pembrolizumab. The registry does not report a trial-specific sample-size justification. The safety figures include deaths, but do not identify their timing, causes, or attribution. These limitations mean the registry record can support descriptive statements about the posted values, but not a complete assessment of efficacy, safety causality, or the assumptions behind the analyses.
15. Related Tutorials
16. Related Statistical Calculators
17. Sources
- ClinicalTrials.gov: Pembrolizumab in Treating Patients With High Risk Oral Intraepithelial Neoplasia — NCT02882282.
- Registry record last updated 2025-10-23; status last verified 2025-08.
- Results first posted 2025-10-23.
- 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.
Continue through the Clinical Biostats statistical pathway
Use this trial's endpoint as a starting point for deeper study of effect measures, confidence intervals, and sample-size planning.
18. Record Summary
This completed, randomized, parallel-group phase 2 study compares observation with pembrolizumab for high-risk oral intraepithelial neoplasia. The registry lists enrollment of 15, with 7 participants in the observation arm and 8 in the pembrolizumab arm. No masking is listed.
The primary outcome is oral cancer-free survival, with a stated timeframe of 51.2 months. The pembrolizumab arm has a reported median of 6.7 months, with a 95% confidence interval from 3.6 to 40.2 months. No median is reported as estimable for observation, and no between-arm analysis is posted. Overall survival is listed as a secondary outcome without a posted result.
The safety record reports no serious adverse events in either arm, other adverse events for 0 of 7 and 6 of 8 participants, and deaths for 1 and 7 participants, respectively. Causes and attribution are not reported. The registry does not provide the endpoint definition, analysis details, or explanation for the unavailable observation-arm median.