Metformin for the Prevention of Oral Cancer in Patients With Oral Leukoplakia or Erythroplakia: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Metformin for the Prevention of Oral Cancer in Patients With Oral Leukoplakia or Erythroplakia (NCT05237960), a randomized phase 2 study. The primary outcome is Histologic Response to Metformin.

2. Clinical Question

The central question is whether oral lesions in participants assigned to extended-release metformin show a histologic response during the study, compared with lesions in participants assigned to placebo. The registry record describes the trial as a study of oral cancer prevention in people with oral leukoplakia or erythroplakia. Its primary endpoint, however, is histologic response in a target lesion, defined by improvement in dysplasia or hyperplasia rather than by a reported count of oral cancers.

This distinction matters when interpreting the prevention question. Histologic changes may be relevant to the study’s stated purpose, but the endpoint description alone does not establish whether oral cancer developed, whether its occurrence differed between arms, or whether any difference was statistically precise. Those outcomes are not reported in the registry record.

3. Trial Design

This is a phase IIb, randomized, double-masked, parallel-group trial sponsored by the University of Arizona. The registry record lists an enrollment of 34 and two study arms: extended-release metformin hydrochloride and placebo. Participants in both arms are described as taking the assigned study treatment by mouth once daily for 24 weeks, unless disease progression or unacceptable toxicity occurs. Both groups also undergo biopsies and blood collections.

The parallel design means the two groups receive their assigned regimens during the study rather than being described as switching between treatments. Randomization is intended, as a general design principle, to help make groups comparable at the start of a trial. The specific randomization procedure and allocation ratio used in the trial are not reported in the registry record. The record identifies the study as double-masked but does not state which roles or individuals were masked, or how masking was maintained.

The primary endpoint is histologic response, assessed up to 24 weeks. The registry defines response as either complete response or partial response, based on changes in dysplasia or hyperplasia in the target lesion. The listed secondary outcomes include clinical response, cell proliferation, changes in C-peptide, glucose, and HbA1C, and plasma metformin concentrations. The record lists 7 outcome measures and 5 analyses, but does not report the detailed analysis methods for these outcomes.

The study start date is 2023-01-12, the primary completion date is 2025-11-07, and the status is active, not recruiting. These registry entries describe the study’s administrative status and planned assessment period; they do not supply a primary-endpoint result. A detailed protocol, the rationale for the enrollment target, and any interim analysis plans are not reported in the registry 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

Analysis population for the primary outcome, as defined in the registry: 29 participants were on study long enough to be evaluable for the primary endpoint

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 Histologic Response to Metformin, measured over Up to 24 weeks.

The primary endpoint is histologic response to metformin, recorded as a count of participants, with an assessment timeframe of up to 24 weeks. The registry defines complete response as complete reversal of dysplasia or hyperplasia to normal epithelium in the target lesion. Partial response means improvement in the degree of dysplasia or hyperplasia. No change means no change in degree, including any outcome that is not classified as complete response, partial response, or progressive disease. Progressive disease means an increase in the severity grade of histology. For the endpoint, histologic response consists of complete or partial response.

Histology describes tissue characteristics seen on examination of a sample, here obtained through biopsy. As a general statistical and clinical-trial concept, a prespecified endpoint gives a study a defined outcome to measure, which helps readers understand what the main analysis is intended to address. In this trial, the endpoint captures changes in the target lesion’s histologic classification rather than a direct count of oral cancer cases.

The registry record says that 29 participants were on study long enough to be evaluable for this endpoint. It does not report how many had complete response, partial response, no change, or progressive disease. It also does not report an effect estimate or a statistical comparison between the treatment arms. Therefore, the endpoint definition and evaluable population are available, but the observed primary-endpoint result is not reported in the registry record.

7. Why No Re-analysis Is Shown

8. Statistical Interpretation

9. Statistical Methods Explained

The registry record gives the trial’s broad design and endpoint definition, but the statistical analysis plan is not reported in the registry record. The concepts below explain how results for a binary response endpoint are commonly interpreted; they should not be read as a description of analyses confirmed for this trial.

What does randomization contribute?

As a general principle, randomization assigns participants to study groups by chance. It aims to reduce systematic differences between groups at the outset, including differences that may not be measured. Randomization does not guarantee that groups will be identical, and it does not remove the need to examine how outcomes were analyzed. The registry record states that this trial is randomized but does not report the randomization sequence, allocation concealment procedures, or exact allocation method.

What does masking mean for interpretation?

Masking is generally used to limit the influence of knowledge about treatment assignment on participant behavior, clinical care, assessment, or analysis. The registry identifies the study as double-masked. The individuals whose knowledge was masked and the procedures used to maintain masking are not reported in the registry record. It is therefore possible to describe the stated design, but not to assess masking success from the information reported.

How should a binary response endpoint be understood?

A binary endpoint groups outcomes into two categories, such as response and no response. This trial’s endpoint definition classifies complete or partial histologic improvement as response. Its other stated categories are no change and progressive disease. The registry record does not report the observed counts in those categories or a rule for handling any ambiguous or unavailable assessment. Without the counts, a comparison of response between arms cannot be interpreted from the posted results.

In general, an effect estimate describes the observed contrast between groups, while a confidence interval describes the range of values compatible with the data under the analysis assumptions. A narrower interval indicates greater precision; a wider interval indicates more uncertainty. Neither an estimate nor a confidence interval for histologic response is reported in the registry record. No primary-endpoint p-value is reported either. A p-value, in general, measures how surprising data would be under a specified null hypothesis and model; it does not measure the probability that a treatment is effective or the size of an effect.

What does the planning illustration represent?

The registry describes planning illustrations based on the observed effect, with a two-sided significance level of 0.05, equal allocation between the arms, and a normal approximation. It says the effect to detect is a stated fraction of the observed effect. These are assumptions for illustrations, not confirmation of the trial’s original design assumptions or its actual analysis plan. The sample-size rationale used to plan the trial is not reported in the registry record. A planning calculation also is not observed power: it cannot establish how likely the completed study was to detect a particular effect after the data were collected.

For general interpretation, a sample-size calculation asks how much information a study would need under stated assumptions to meet a chosen statistical criterion. Such calculations depend on the effect assumed, outcome variability or event frequency, allocation, and analysis method. If the assumed effect is based on an observed estimate, that estimate may itself be uncertain. Planning illustrations should therefore be read as conditional scenarios rather than as evidence about the outcome. Numerical planning results are not reported in the registry record’s statistical details.

Why can binary outcomes need special care?

With a binary endpoint, each participant contributes to a category such as response or no response. The interpretation depends on the number of observed outcomes in each arm and on the analysis population. As a general point, small numbers of events can leave substantial uncertainty around a group contrast. The enrollment and evaluable-population counts alone do not determine the response effect; the outcome counts and analysis are also needed.

Some count outcomes can show over-dispersion, meaning that variability is greater than a simple model expects. This matters because a model that underestimates variability can make uncertainty look smaller than it is. Over-dispersion is a general concept and is not identified as a feature of this binary primary endpoint. The registry record does not report a model choice, adjustment variables, or a method for assessing model assumptions.

Multiple outcomes and analyses can also complicate interpretation. If many hypotheses are tested, the chance of at least one apparently notable result can increase unless the analysis plan addresses multiplicity. The registry lists secondary outcomes and analyses, but a prespecified testing hierarchy or multiplicity adjustment is not reported in the registry record. No confirmatory statistical interpretation should be inferred from the number of outcomes alone.

Finally, an analysis population defines which participants contribute to a particular estimate. The record identifies 29 participants as evaluable for the primary endpoint but does not describe whether the main analysis follows randomized assignment, includes only participants with observed endpoint data, or uses another approach. Those methods can answer different questions. The analysis population, missing-data procedure, and sensitivity analyses are not reported in the registry record.

10. Multiplicity, Missing Data and Analysis Populations

The registry record lists 7 outcome measures and 5 analyses, including the primary histologic response endpoint and secondary outcomes involving clinical response, cell proliferation, metabolic measures, and plasma metformin concentrations. A list of outcomes does not show how their analyses were ordered or interpreted. A prespecified hierarchy, adjustment for multiple comparisons, or a distinction between confirmatory and exploratory analyses is not reported in the registry record.

Multiplicity is a general statistical issue that arises when several outcomes, comparisons, or analyses are examined. Without a plan for handling multiple tests, an isolated result can be harder to interpret as confirmatory. The registry’s listed outcomes do not by themselves establish that a particular adjustment was or was not used; the statistical plan is not reported in the registry record.

The primary endpoint population is described as 29 participants who were on study long enough to be evaluable. The number of participants with missing endpoint assessments, reasons for missingness, and the method used to handle unavailable outcomes are not reported in the registry record. The record also does not state whether the primary analysis included all randomized participants, only evaluable participants, or another population. A complete-case analysis, an imputation approach, and other methods can rely on different assumptions, so the missing-data method matters to interpretation.

There is no reported sensitivity analysis showing how the primary comparison changes under alternative assumptions about missing outcomes. The protocol’s prespecified analysis population and approach to missing data are not reported in the registry record. These omissions mean that the primary endpoint’s analytic denominator and the robustness of a future or posted estimate cannot be assessed from the information reported.

11. 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: oral pain (5.9%).

The registry reports safety counts for two categories, serious adverse events and other adverse events, with 17 participants at risk in each arm for each category. In the extended-release metformin arm, 1 participant was affected by a serious adverse event and 15 were affected by other adverse events. In the placebo arm, 0 participants were affected by a serious adverse event and 13 were affected by other adverse events. No deaths are listed in either arm.

These are counts of participants affected, not a complete description of the type, duration, severity, or clinical course of every event. The registry record lists oral pain at 5.9% among the top serious adverse-event terms. The information reported does not give an arm-specific count for oral pain, so that term should not be assigned to one arm based on this listing.

As a general principle, safety counts are descriptive and should be interpreted in light of the number of participants observed and the way events were collected. The registry information does not report a statistical comparison of adverse-event rates or provide detail about event timing, recurrence, or attribution. The counts should not be interpreted as proof that an event was caused by, or unrelated to, the assigned study treatment.

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

The registry record establishes that a randomized, double-masked, parallel phase IIb trial compares extended-release metformin with placebo in participants with oral leukoplakia or erythroplakia. It defines histologic response in terms of complete or partial improvement in dysplasia or hyperplasia in a target lesion, assessed up to 24 weeks. It also states that 29 participants were on study long enough to be evaluable for the primary endpoint and reports adverse-event counts by arm.

Those details describe the question and the available reporting, but they do not demonstrate a primary-endpoint result. Histologic response counts by category, an effect estimate comparing arms, a confidence interval, and a p-value are not reported in the registry record. The information therefore cannot establish whether the groups differed in histologic response or how precisely any difference was estimated.

The trial’s stated purpose concerns prevention of oral cancer, while its primary endpoint is histologic response in a target lesion. The endpoint may provide information about tissue changes relevant to the research question, but the endpoint definition alone is not evidence that oral cancer was prevented. Oral cancer incidence or a between-arm comparison for that outcome is not reported in the registry record. Nor should the absence of a reported result be interpreted as evidence of no effect.

The listed adverse-event figures provide a separate, descriptive view of reported safety events. They do not answer the primary efficacy question. The registry record does not report details needed to compare event timing, severity beyond the listed categories, or causal attribution.

As a general statistical principle, conclusions depend on both the estimated effect and its uncertainty, as well as on the analysis population and assumptions. Those elements are not reported for the primary endpoint. The most defensible interpretation is therefore limited: the registry describes a planned comparison and outcome definitions, but the posted information does not permit a conclusion about comparative histologic response or cancer prevention.

13. Why This Trial Matters Statistically

This trial is statistically informative as an example of the distinction between a research objective and the outcome actually used to assess it. The stated objective is oral cancer prevention among people with oral leukoplakia or erythroplakia. The primary endpoint is a histologic response in a target lesion, defined through changes in dysplasia or hyperplasia. Readers need to keep those concepts separate: a tissue-response endpoint is not itself a reported cancer-incidence result.

The design also illustrates why a randomized comparison needs outcome data to support an inference. The registry identifies a metformin arm and a placebo arm and describes the study as randomized and double-masked. Yet without the response counts and comparative estimate, the design description alone cannot tell readers whether outcomes differed or how precisely any difference was measured.

The evaluable population is relevant because the record identifies 29 participants as having been on study long enough for the primary endpoint. Reporting the endpoint denominator, reasons for non-evaluability, and the analysis population would help readers assess how the observed participants relate to the enrolled trial population. Those details are not reported in the registry record.

Finally, the presence of multiple outcome measures makes transparent analysis planning important. Different outcomes can address distinct biological or clinical questions, and a clear prespecified plan helps readers distinguish primary evidence from exploratory findings. The registry lists secondary outcomes and analyses, but the analysis plan and multiplicity approach are not reported. These features make the record a useful reminder that design information, endpoint definitions, completed analyses, and interpretable results are separate parts of the evidence.

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 for interpreting efficacy is that the registry record does not report the primary-endpoint response counts or an arm-to-arm effect estimate. It therefore does not show how many participants had complete response, partial response, no change, or progressive disease, and it does not report uncertainty around a comparison. A p-value or confidence interval for histologic response is not reported in the registry record.

The evaluable population is stated as 29 participants, but the reasons other participants were not evaluable and the handling of unavailable assessments are not reported. The record also does not specify the primary analysis population, analysis model, covariate adjustment, or sensitivity analyses. These details would be needed to understand how the endpoint data were translated into a comparison and how dependent that comparison might be on analytical assumptions.

The listed primary endpoint is a histologic response assessed up to 24 weeks. The registry information does not report an oral cancer incidence result or another direct comparative cancer outcome. As a result, the prevention objective cannot be evaluated from the endpoint definition alone. Follow-up beyond the stated endpoint timeframe and any later outcomes are not reported in the registry record.

The record identifies the study as randomized and double-masked but does not describe the randomization method, the masked roles, or how masking was maintained. It lists secondary outcomes and analyses without reporting the statistical plan or multiplicity approach. The sample-size rationale for the trial is also not reported in the registry record. The planning assumptions described for illustrations should not be treated as the trial’s original design justification.

Safety information is limited to affected-participant counts, deaths, and a listed serious adverse-event term. Event-level details, timing, duration, attribution, and a statistical comparison are not reported in the registry record. These limits constrain interpretation of both benefits and harms from the posted information.

15. Related Tutorials

16. Related Statistical Calculators

17. Sources

• ClinicalTrials.gov: Metformin for the Prevention of Oral Cancer in Patients With Oral Leukoplakia or Erythroplakia — NCT05237960.
• Registry record last updated 2026-09-28; status last verified 2026-09.
• Results first posted 2026-09-02.
• 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 record describes NCT05237960, titled “Metformin for the Prevention of Oral Cancer in Patients With Oral Leukoplakia or Erythroplakia,” and identifies the study as M4OC-Prevent 2.0, a phase IIb trial. It is randomized, double-masked, and parallel, with an enrollment of 34. The sponsor is the University of Arizona. The listed study start date is 2023-01-12, the primary completion date is 2025-11-07, and the status is active, not recruiting.

The two arms receive extended-release metformin hydrochloride or placebo by mouth once daily for 24 weeks, with biopsies and blood collections in both groups. The primary endpoint is histologic response up to 24 weeks; 29 participants are described as evaluable. The response-category counts and comparative primary-endpoint results are not reported in the registry record.

Secondary outcomes include clinical response, cell proliferation, changes in C-peptide, glucose, and HbA1C, and plasma metformin concentrations. Safety reporting lists serious and other adverse-event counts by arm and no deaths in either arm. The detailed statistical plan, missing-data methods, analysis populations, and multiplicity approach are not reported in the registry record.