UAB 0718 - Phase II Trial to Assess Target Oral Therapy as Adjuvant Chemoprevention in High-Risk Head and Neck Cancer: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for UAB 0718 - Phase II Trial to Assess Target Oral Therapy as Adjuvant Chemoprevention in High-Risk Head and Neck Cancer (NCT00570232), a na phase 2 study. The primary outcome is Number of Participants Demonstrating the Safety and Tolerability of Long Term Erlotinib Treatment.

2. Clinical Question

The clinical question is exploratory: what safety and tolerability experience was recorded during erlotinib treatment in this population? The trial title also frames the intervention as adjuvant chemoprevention, while the brief summary describes the purpose as assessing whether erlotinib would control cancer that had returned after salvage surgery and radiation. Those descriptions indicate the study's aims; they are not evidence that the treatment controlled cancer.

The registry lists survival at 12 Months and 24 Months as a secondary outcome. However, no survival results are included in the registry record described for this analysis. The record also does not report a treatment-effect estimate against a comparator. Accordingly, the available information supports a description of the recorded safety counts, not an estimate of comparative benefit or harm.

3. Trial Design

This was a Phase II, single-group study sponsored by the University of Alabama at Birmingham. The registry lists enrollment of 31 and describes the study as completed. Its start is recorded as 2007-12 and its primary completion as 2014-12. The listed condition is Head and Neck Cancer.

The intervention was erlotinib, with the arm description stating that all patients would be prescribed erlotinib 150mg daily. Because the design has one group, there is no concurrent control group against which to compare outcomes. Allocation is listed as NA, and masking is listed as NONE. These design features matter when interpreting both safety and any proposed assessment of cancer control: changes observed during follow-up cannot be separated from the course of disease, other care, or other influences using a randomized comparison in this study.

The registry's stated purpose was to determine whether erlotinib would be effective in controlling cancer that had returned after treatment with salvage surgery and radiation, and to determine the drug's effects, good and/or bad. The primary outcome is framed as the number of participants demonstrating safety and tolerability of long-term treatment. The timeframe is 12 - 24 months. Survival at 12 Months and 24 Months is listed as a secondary outcome.

The registry record does not report detailed eligibility criteria, the intended treatment duration for each participant, adherence, dose modifications, or the schedule for collecting outcome information. It also does not report a statistical analysis plan or outcome analyses. These omissions mean that the design can be described from its group structure and stated outcomes, but the planned analysis and the details of implementation cannot be reconstructed from the record.

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 Number of Participants Demonstrating the Safety and Tolerability of Long Term Erlotinib Treatment, measured over 12 - 24 months.

The primary outcome is titled “Number of Participants Demonstrating the Safety and Tolerability of Long Term Erlotinib Treatment.” It is measured in participants, with the parameter described as a number. The accompanying description says it counts participants who had the most frequently observed undesirable effects after exposure to study drug. The stated timeframe is 12 - 24 months.

A count of participants can communicate how many people experienced an outcome, rather than how many separate episodes occurred. This distinction is useful in safety reporting: a participant may have more than one undesirable effect, while a participant-level count generally treats each person as the unit for the outcome. The registry wording does not specify how the primary outcome classified severity, how “most frequently observed” was defined, or whether the endpoint combined particular effects. Those details are not reported in the registry record.

Safety and tolerability are relevant in a study of ongoing drug exposure because they describe the experience of participants during treatment and follow-up. A count alone, however, does not show how events compare with what would have occurred without the drug, nor does it establish that the drug caused an event. The single-group design also limits comparative interpretation.

The registry entries include adverse-event group counts, but do not provide a separate reported result for the named primary endpoint. It is therefore important not to treat the adverse-event counts as a complete analysis of the primary outcome or to infer an endpoint definition beyond its stated description.

7. Why No Re-analysis Is Shown

8. Statistical Interpretation

9. Statistical Methods Explained

A single-group design describes outcomes in one treated group. It can provide a record of events and follow-up outcomes among participants, but it does not directly estimate how those outcomes would differ under another treatment or no treatment. In a randomized comparison, allocation is used to create groups that can be compared under a defined analysis. This study has no such comparison group: allocation is listed as NA, the model is single-group, and masking is NONE. The consequences are interpretive rather than a reason to disregard the descriptive information: observed events can be counted, but causal attribution and comparative effects are not established.

What does a participant count represent?

A participant count records people meeting an outcome definition. It differs from an event count, which may include repeated occurrences in the same person. For safety summaries, a participant-level count can make the proportion of affected people easier to understand, while an event-level summary can convey recurrence or total burden. The primary outcome is described as a number of participants, and the adverse-event group reports participants at risk and participants affected. The registry record does not specify the detailed rules used to deduplicate events, define the primary endpoint, or classify each event.

Counts must be interpreted with their denominators. “At risk” identifies the group used as the denominator for the reported adverse-event category. In this record, the serious and other adverse-event categories each list 31 at risk. The corresponding affected counts are 9 and 31. These values should be quoted as reported rather than transformed into additional percentages. A denominator does not by itself reveal whether all participants had equivalent observation time, whether follow-up was complete, or whether exposure was similar.

Why does over-dispersion matter?

In general, a statistical model for counts may assume a particular relationship between the average count and its variability. Over-dispersion means the observed variability is greater than that model expects. If it is ignored, estimated uncertainty can be too small and inferential results can appear more precise than warranted. This is a general issue in count analysis, not a claim about this study: the registry entries do not report an over-dispersion assessment, a count model, or an analysis of event rates. The primary endpoint is a participant count, and the registry entries do not support reconstructing a model for repeated events.

What do confidence intervals and significance tests do?

In general, a confidence interval expresses the uncertainty around an estimate under a specified statistical procedure and its assumptions. A narrow interval indicates greater precision under those assumptions; a wide interval indicates less precision. A significance test evaluates compatibility between data and a specified null hypothesis, also under assumptions. Neither an interval nor a p-value measures clinical importance on its own, proves causation, or corrects weaknesses in design. No confidence intervals, p-values, or outcome analyses are reported in the registry entries for this trial, so no such inference can be made from the available results.

For an outcome with a binary status, a group may be described by participants with and without the event. Comparative methods would require a defined comparison and appropriate assumptions. Here, there is no comparator arm, and no unadjusted or adjusted treatment contrast is reported. A rate or proportion in the treated group can describe observed experience, but it cannot reveal the counterfactual outcome that the same participants would have had under another strategy.

How should planning sample sizes be understood?

In general, a sample-size calculation is a design tool. It depends on assumptions about the effect to detect, variability, allocation, significance threshold, and desired power. A planning illustration is not an observed result and cannot show what the trial actually found. The stated illustration assumptions refer to a two-sided significance level of 0.05, equal allocation between two arms, a normal approximation, and an effect specified as a fraction of the observed effect. They do not describe the actual single-group design. The registry entries do not report the illustration's numerical outputs or the effect estimate on which it was based. No trial-specific sample-size justification is reported in the registry record.

Power is a design probability under a particular set of assumptions. It is not a measure of whether a completed study has established a result, and retrospective calculations based on an observed effect do not substitute for reported estimates and uncertainty. Any sample-size figures associated with planning should therefore be read as hypothetical illustrations for a comparison, not as evidence that this trial was designed or powered in that way.

Interpretation also depends on outcome definition and time. The primary timeframe is 12 - 24 months, while the secondary survival outcome names 12 Months and 24 Months. These labels do not describe the analysis method, censoring rules, or handling of participants without complete follow-up. For time-to-event outcomes, general methods account for the time participants are observed and whether an event has occurred by their last observation. The registry entries do not report a time-to-event analysis, event counts, or censoring information. Thus, the listed survival outcome cannot be interpreted as an estimated survival probability in the registry record alone.

10. Multiplicity, Missing Data and Analysis Populations

Multiplicity arises when a study assesses several outcomes, time points, or statistical tests. Testing many questions can increase the chance of at least one apparently notable result by chance unless the analysis plan addresses that issue. The registry lists a primary safety and tolerability outcome and a secondary survival outcome at 12 Months and 24 Months. It also states that there are 3 outcome measures. The registry entries do not describe how these outcomes were prioritized beyond the primary and secondary labels, whether any multiplicity adjustment was planned, or which analyses were intended.

The record reports 0 analyses. That means the available the registry entries do not include outcome-analysis results to evaluate; it does not mean that no analysis was ever conducted outside the registry record. Statistical analysis plans, model specifications, prespecified contrasts, and any adjustment for multiple comparisons are not reported in the registry record.

Missing-data handling is also not described. The record does not say whether follow-up was complete, how withdrawals or unavailable measurements were treated, or whether any imputation or sensitivity analysis was used. The adverse-event group lists 31 at risk for both categories, but those denominators do not resolve missingness for the primary endpoint or secondary survival outcome.

The analysis population is not defined beyond the reported group counts. It is not reported whether outcomes were analyzed for all enrolled participants, only those who began treatment, or another population. Without these details, readers cannot assess potential bias from exclusions, incomplete follow-up, or selective outcome reporting.

11. Safety Results

The registry reports serious adverse events as affected participants divided by participants at risk.

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: pneumonia leading to respiratory failure (3.2%), bleeding from throat (3.2%), stridor (3.2%), bell's palsy (3.2%), pneumonia (3.2%).

The Tarceva group reports 31 participants at risk for serious adverse events and 9 participants affected. For other adverse events, the group reports 31 at risk and 31 affected. These counts describe the registry's categories; they do not establish that every event was caused by the study drug.

The listed serious-event terms are pneumonia leading to respiratory failure, bleeding from throat, stridor, Bell's Palsy, pneumonia, and cardiac stents placed. Each is reported with a percentage of 3.2%. The registry entries do not give the number of occurrences for each term, whether terms refer to distinct participants, or whether a participant could have more than one listed event. The percentages should therefore be read as reported alongside the terms, not used to infer a total or reconstruct participant-level data.

Serious adverse events and other adverse events are separate reporting categories. Their counts should not be combined. The record does not state the classification criteria, timing, duration, relationship assessment, or clinical course for these events. Nor does it provide a comparison group to estimate how often similar events might have occurred without erlotinib. The figures are descriptive safety information, not a causal estimate of drug-related risk.

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

The registry record documents a completed, single-group Phase II study of erlotinib in a high-risk head and neck cancer context. Its primary outcome is framed around the number of participants demonstrating safety and tolerability of long-term treatment. The record also gives adverse-event counts for the Tarceva group, including serious adverse events for 9 of 31 participants and other adverse events for 31 of 31 participants. These data describe recorded experience in that group.

They do not establish whether erlotinib controlled cancer, prevented recurrence, or changed survival. Although the study's stated purpose includes cancer control and its secondary outcomes include survival at 12 Months and 24 Months, the registry entries do not report results for those outcomes. No analysis estimate, confidence interval, or comparative result is available in the registry entries described for this analysis.

The lack of a control group is central. A single treated group cannot show what would have happened to comparable participants under a different approach. Events may reflect the underlying cancer, previous salvage surgery and radiation, other care, the study drug, or multiple influences. The record does not provide an analysis that separates those explanations.

The counts also do not describe each participant's treatment exposure, event timing, event duration, or completeness of follow-up. The record does not report detailed outcome definitions or missing-data methods. Those limits prevent a more refined assessment of safety patterns or outcome precision.

Accordingly, the appropriate interpretation is descriptive and bounded: the registry reports the listed participant counts and adverse-event terms in a single-group study. It does not establish a treatment effect, prove that an event was caused by erlotinib, or provide comparative evidence of benefit or harm. These limits do not negate the value of recording safety experience; they define what conclusions the record can support.

13. Why This Trial Matters Statistically

This study illustrates why design determines the kind of statistical question that can be answered. A single-group trial can document what happened among participants receiving an intervention, including participant counts for safety outcomes. Without a comparator, however, it cannot estimate a treatment contrast. That distinction is essential when a study's stated aims include both safety and cancer control.

The record also shows why outcome labels and outcome results should not be conflated. The primary outcome is titled around safety and tolerability, and the secondary outcome names survival at 12 Months and 24 Months. Listing an outcome identifies a question of interest; it does not itself provide an estimate. Here, the registry entries include adverse-event counts but no outcome analyses, and they do not report survival results.

Safety data require careful attention to the unit of reporting. Participant counts are not necessarily event counts, and the record does not explain whether a person could contribute to multiple listed terms or categories. Reporting the denominator alongside the affected count helps describe the observed group, but it does not solve questions about exposure duration, follow-up completeness, or attribution.

Finally, planning calculations and completed-trial evidence serve different purposes. Sample-size illustrations depend on assumptions about a hypothetical comparison and an effect to detect. They cannot replace a reported analysis, nor can they convert a single-group design into a comparative one. This distinction helps prevent design assumptions from being mistaken for findings.

For clinical researchers and biostatisticians, the case underscores a basic discipline: separate the stated objective, the design, the recorded observations, and the conclusions supported by those observations. The registry's details support a cautious description of adverse-event reporting, while leaving comparative effectiveness, outcome precision, and analysis choices unresolved.

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 most important limitation is the absence of a control group. The study is recorded as single-group, with allocation NA and masking NONE. Consequently, the reported adverse-event counts cannot be compared with a concurrent group receiving another treatment or no study drug. The record cannot establish whether an event was caused by erlotinib or estimate a comparative treatment effect.

The primary outcome's wording does not provide a full operational definition. It refers to participants with the most frequently observed undesirable effects after study-drug exposure, but does not identify a complete event list, grading system, attribution rule, or method for combining effects. The adverse-event counts therefore should not automatically be treated as the primary endpoint result.

Further details about enrollment and follow-up are absent. The record does not report eligibility criteria, participant characteristics, actual treatment exposure, adherence, dose changes, duration on treatment, or completeness of outcome ascertainment. The reported at-risk counts do not settle these questions for all outcomes.

The registry entries list survival as a secondary outcome but do not report survival results, time-to-event analyses, censoring methods, or uncertainty estimates. They also report 0 analyses, and do not provide a statistical analysis plan, missing-data procedures, or a definition of analysis populations. Multiplicity handling is not reported in the registry record.

Finally, the registry lists several serious adverse-event terms with a percentage for each but does not provide event-level context or clarify overlap among participants and terms. These omissions constrain interpretation of both safety patterns and the study's stated cancer-control objective.

15. Related Tutorials

16. Related Statistical Calculators

17. Sources

• ClinicalTrials.gov: UAB 0718 - Phase II Trial to Assess Target Oral Therapy as Adjuvant Chemoprevention in High-Risk Head and Neck Cancer — NCT00570232.
• Registry record last updated 2015-05-21; status last verified 2015-05.
• Results first posted 2015-05-21.
• 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.

18. Record Summary

The registry describes NCT00570232, a completed Phase II, single-group study sponsored by the University of Alabama at Birmingham. It lists Head and Neck Cancer as the condition and enrollment of 31. The study start is recorded as 2007-12, with primary completion at 2014-12.

The intervention was erlotinib, identified as Tarceva, prescribed at 150mg daily. The primary outcome concerns the number of participants demonstrating safety and tolerability of long-term treatment, with a timeframe of 12 - 24 months. Survival at 12 Months and 24 Months is listed as a secondary outcome.

For the Tarceva group, serious adverse events were reported for 9 of 31 participants at risk, and other adverse events for 31 of 31. The record lists serious-event terms with percentages of 3.2%. It reports no analyses. Detailed analysis methods, survival results, missing-data handling, and a comparative treatment effect are not reported in the registry record.